EXPANDING YOUR COMFORT ZONE – the Effects of Artistic and Cultural
Intervention on the Workplace
A Study of AIRIS 2005-2008
(Including Genklang Vara 2006-2008)
Institute for Management of Innovation and Technology
Michael Eriksson
February 2009
Translation: Evert Wängberg, TILLT AB
2
Summary
The objective of Artists in Residence, AIRIS, is establishing a deep and far reaching collaboration
between Culture and Industry operating within private and public business sectors, and focussing
upon the creative processes that are set in motion whenever a professional artist encounters people
at a workplace.
Before the 2005 AIRIS Project, the Institute for Management of Innovation and Technology (IMIT)
was signed up by Skådebanan Västra Götaland for the task of performing a quantitative study of the
specific effects of artistic and cultural intervention on the workplace that an AIRIS run would
imply. We have a broad experience of change studies within all kinds of business sectors. Based
upon conversation with half of the companies that participated in AIRIS 2004, we have selected an
array of tools from the field of established and tested instruments.
Quantitative data were assembled by a survey at project launch as well as project termination. By
constructing the surveys using established indexes of prognostic validity we have obtained good
comparison data. We used the following analytic parameters to captivate the effects of the
intervention:
Organizational Climate for Creativity and Innovation, (Ekvall, 1996)
Patterns of interaction in work groups, (Kylén, 1999)
Experimenting
Managing Complexity and Uncertainty, (Adler, 1999)
The Outlook on Planning and Efficacy, (Adler, 1999)
Change Stategies, (Norrgren, Hart, Schaller, 1996)
Average short term sick leaves at the better part of workplaces showed a more positive trend in the
Genklang Project compared to average in the Vara Municipality the year of the AIRIS run.
However, subsequently to project termination, the trend does not seem permanent.
The significant changes described in the quantitative study contain expressions such as “meeting
new people and getting a new outlook on my work” and “breaking conventional patterns”. These
expressions may be regarded as next of kin to characteristics often associated with artists, that is,
the capability of working from alternative perspectives and managing things that evade prediction.
Other trends from the quantitative material involve “decreased resistance to change” and attitudes
toward leadership, “a good boss sees possibility and adapts his business accordingly”. Such trends
signal increased inclination to change, both directly (decreasing resistance) and indirectly
(embracing possibility). The supplementary quantitative material repeats a multi levelled experience
of increased cooperation and cohesion as well as better workplace atmosphere.
Through the AIRIS Project we may say executives and co-workers at the participant workplaces
received an impulse that forced them out of their comfort zone, or in several cases, expanded it, thus
potentially making for an expansion of change and innovation space. In order to develop their
innovation processes organisations need, first, enhance their capability of embracing new
knowledge by identifying and comprehending its value-generating potential, and second, develop
new concepts and commercial innovations by experimenting and generating potentially value-
generating knowledge. The integration of the two processes that embrace and generate knowledge
of potential value, is seen as crucial for making effective and innovative product development
possible.
3
Table of Contents
Summary 2 Table of Content 3 AIRIS 4 Quantitative Results 5
Organizational Climate for Creativity and Change 5 Patterns of interaction in work groups 5 Creative vs Efficacy Logic 5 Project Effects and Contributions to Improving Indicators 6 Backdrop Variables 7 T Polls: Pre–Post 8 Correlations of Index–Post 9
Climate Index 9 Efficacy and Creativity Indexes 9 Defensive Index 9 Effect-1 and Effect-2 Indexes 10
Climate Index 11 Efficacy Logics Index 12 Creativity Logics Index 13 Defensive Action Patterns Index 14 Effect-1 and Effect-2 Indexes 15 Conclusions 17
Sick Leave Results 18 Conclusions 20
Qualitative Results 21 What Did the Participation Achieve? 21 PR Value According to Skådebanan’s assessment 22 Housing Company Bostadsbolaget’s Estimations 22
Conclusions 23 References 25
4
AIRIS
The objective of Artists in Residence, AIRIS, is establishing a deep and far reaching collaboration
between Culture and Industry, operating within private and public business sectors, and focussing
upon the creative processes that are set in motion whenever a professional artist encounters people
at a workplace. AIRIS is run by Skådebanan Västra Götaland Kultur och Arbetsliv (today Tillt
AB,www.tillt.se).
Behind AIRIS stands the urge of utilising a special competence that involves knowledge of the
creative processes and experiences of the unpredictable that is present in the culture sector, and
using that competence in change and development work at a workplace. During this process the
artist offers new and alternative approaches and modes of thinking.
The AIRIS artists are hand-picked by Skådebanan1 and present at the workplace on average one day
a week during a total of eight months. The artist and a project group will jointly determine the areas
on which they are going to work prior to jointly forming a concrete action plan.
Before the 2005 AIRIS Project, the Institute for Management of Innovation and Technology (IMIT)
was signed up by Skådebanan Västra Götaland with the task of performing a quantitative study of
the specific effects of artistic and cultural intervention on the workplace that an AIRIS run would
imply. We have broad experiences of doing studies on change in all kinds of business. Based upon
conversation with half of the companies that participated in AIRIS 2004, we have selected an array
of tools from the field of established and tested instruments.
Over a span of four years we have had the privilege of following the participant workplaces during
their AIRIS project as well as the workplaces for two years under the sister project in Vara, known
as Genklang2. Since 2006 we also have supplemented the quantitative data assembly with a limited
qualitative data assembly done mainly by interviews at participant workplaces, but we also shared
Skådebanan’s evaluating meetings with workplace participants. Since 2007 we have collected data
concerning the sick leave trend developments at the participation workplaces.
1 The name was taken from the Shakespeare comedy As You Wish and means The Stage.
2 Genklang means Resonance.
5
Quantitative Results
Quantitative data were assembled by surveys done at both project launch and project termination.
By constructing the surveys using established indexes of prognostic validity we have obtained good
comparison data. The following analytic parameters for catching the effects of the intervention were
used:
Organizational Climate for Creativity and Innovation, (Ekvall, 1996)
Patterns of Interaction in Work Groups, (Kylén, 1999)
Experimenting
Managing Complexity and Uncertainty, (Adler, 1999)
The Outlook on Planning and Efficacy, (Adler, 1999)
Change Stategies, (Norrgren, Hart, Schaller, 1996)
Organizational Climate for Creativity and Change
This instrument was developed during a research programme in Sweden during the 1980s on the
organisation conditions that stimulate or hinders creativity and innovation. The instrument
originally consists of fifty questions spanning ten different dimensions based upon several major
factor analyses. These are: Challenge – involvement and feelings for the organisation and its
objectives; Freedom – the kind of behavioural independence that the people in the organisation
maintain; Idea-Support – how new ideas are received; Trust – the emotional security offered;
Dynamism – the dynamics within the organisation; Playfulness – the prevailing level of ease;
Debate – the extent to which meetings, clashes of views, ideas, alternative experiences, and
knowledge occur; Conflict – the extent to which emotional tensions, as opposed to conceptual
tension, occur within the organisation; Risk Taking – the inclination to tolerate insecurity within the
organisation; Idea-Time – The amount of time allocated and used for working out new conceptions.
Mats Sundgren, in collaboration with Göran Ekvall, developed a reduced instrument made of ten
questions – one for each dimension and used in several studies (Sundgren 2004).
Patterns of interaction in work groups
Researching interaction patterns in workgroups Sven Kylén argued in favour of focussing on
workgroup interaction patterns with the objective of learning to reduce defensive blocking
behaviour in favour of developing the support of offensive behaviour. Kylén developed an
instrument that measures the kind of behaviour of workgroups that impacts learning, capability of
change, and efficacy. It contains twenty-five variables, out of which fifteen are defensive, and ten
are offensive. Defensive interaction patterns are characterised by seclusion, non-disputability, and
inconsistency between word and action. Lack of insights into the bigger picture and egocentricity
are regarded as crucial interaction factors. Offensive interaction patterns, on the other hand, are
characterised by receptiveness to new ideas, and make improvements of the interplay between
group members in order to develop both labour methods and products/services. In collaboration
with Kylén a reduced instrument was developed that overlap with Ekvall’s ten dimensions (see
above).
Creative vs. Efficacy Logics
Studies of companies that have a high level of competition power reveal that they also have a clear
focus on innovation, creative processes, and renewal. Increasing competition and concentration on
cost cuts over the last decades have resulted in one-sided focus upon the Logics of Efficacy. Most
organisations today are not particularly adept at discovering, choosing, and embracing new options.
Organisations are often stuck in historical success formulas and strong processes, and wanting
Creative Logics.
6
For many enterprises the presence and development of Creative Logics represent a basic condition
for commercial innovation and a fundament for sustainable business, both from trade and human
perspectives. Creative Logics has to do with organised ingenuity and is a mark of creativity in the
context of business and its collaborating capacity with the rest of the organisation. A great part of
the knowledge body ignores the context in which creativity arose and, because of this, it is useless.
Creative Logics means thinking outside the frameworks and question what is. This is similar to
what art and culture do. Efficacy Logics, on the other hand means effective execution of existing
things, including improvements that seldom involves innovative development leaps. At the same
time, sustainable and innovative growth demands a balance between guiding logics, creative, and
efficacy logics. It is not a matter of either or, it is a matter of both.
EFFICACY LOGICS CREATIVE LOGICS
Aiming at minimising deviation from plan and budget Plan and budget are necessary; however, more
important are continual experimenting and the
accumulated result
Uncertainty to be minimised by rigorous planning Uncertainty perceived as a necessary condition for
developing new concepts and options
Complexity is best managed by breaking it down Complexity generates many overall views
Efficiency is achieved by functional specialisation and
minimised interdependence of workgroups
Boundary crossing meetings, combining competence
and perspective are necessary conditions for success
Decisions are based upon predetermined quantitative
economic data
Decisions are based upon the results of experimenting
Figure 1 Comparison of Efficacy Logics and Creative Logics
Project Effects and Contributions to Improving Indicators
The effects of more thorough change projects on several of the indicators are often delayed one to
two years after project termination. However, we have found that the instrument used in the present
study often will indicate areas of improvement, even while indicating nothing about its ultimate
potential, i.e. the magnitude of change. The outcome shows appreciation from the participants in
connection with the artist’s termination of the short year at the workplace.
Two question categories were used, of which the first was the degree to which you believe the
project would contribute to any the following:
Better physical work environment
Better working atmosphere
Improved cooperation
Enhanced work involvement
Enhanced work quality
More varying tasks
Increased information exchange
Increased overall view
The second question category was the extent to which the project would contribute to change
concerning any of the following indicators:
Productivity
Quality
Cost levels
Sick leaves
Stress levels
7
In order to draw more generalised conclusions from the quantitative outcome we estimated that
observations from twenty-five to thirty workplaces would do, and that is what we have. This report
is based upon responses from thirty-six workplaces, in total of 1,094 polls. The average post poll
frequency was 76 percent.
Statistical processing and interpretation were done in collaboration with Professor Joseph Schaller
at the Department of Psychology, Göteborg University, and Professor Flemming Norrgren at
Management of Organisational Renewal and Entrepreneurship (MORE), Chalmers University of
Technology.
Backdrop Variables
Respondents
Gender
Female 70 %
Male 29 %
No information 1 %
Age
– 30 11 %
31 – 40 23 %
41 – 50 30 %
51 – 60 27 %
61 – 6 %
No information 3 %
Capacity
Manager 11 %
Worker 87 %
No information 2 %
Ethnicity
Nordic, incl. Swedish 88 %
Extra-Nordic 2 %
No information 10 %
This variable was missing 2005. The account is based on 80 percent of the collected responses.
8
T Polls: Pre–Post
Below are the average values of Pre–Post Polls respectively, question by question. There are four
indexes at the end of the table dealing with Climate, Efficacy, Creativity, and Defensive (for more
information about index, see next section).
Poll N MEAN STD SE
Q 1: Challenge Pre 553 5.6311 1.05873 .04502 Post 443 5.5824 1.13734 .05404 Q 2: Idea support Pre 554 5.0776 1.15496 .04907 Post 442 5.1335 1.09537 .05210 Q 3: Trust Pre 553 5.2333 1.24739 .05304 Post 443 5.1919 1.17928 .05603 Q 4: Freedom Pre 551 5.0708 1.13637 .04841 Post 441 5.0813 1.09408 .05198 Q 5: Playfulness Pre 547 5.1024 1.23759 .05292 Post 443 5.0726 1.20068 .05705 Q 6: Debate Pre 549 4.4062 1.25595 .05360 Post 444 4.4054 1.21969 .05788 Q 7: Risk Taking Pre 551 4.8276 1.03276 .04400 Post 439 4.7813 1.05878 .05053 Q 8: Dynamical Pre 551 4.7114 1.14027 .04858 Post 442 4.8054 1.20210 .05718 Q 9 :Conflict Pre 550 3.1891 1.48749 .06343 Post 444 3.3536 1.40740 .06679 Q 10: Idea Time Pre 548 3.8394 1.15829 .04848 Post 444 3.8514 1.15619 .05487 Q 11: Focused and Pre 550 5.3527 .87837 .03745 Effective Post 444 5.4302 .92719 .04400 Q 12: New Concepts Pre 549 4.4772 1.12452 .04799 Post 444 4.5360 1.03704 .04922 Q 13: New People Pre 550 3.9018 1.38078 .05888 Post 443 4.1174 1.39804 .06642 Q 14: Unconventional Pre 540 3.5167 1.17239 .05045 Post 436 3.6927 1.13949 .05457 Q15: Reflections Pre 550 4.1745 1.13340 .04833 Post 443 4.1219 1.10676 .05258 Q 16: Leaning From Pre 549 4.6612 1.06793 .04558 One Another Post 444 4.6081 1.13612 .05392 Q 19: Clear Planning Pre 549 5.1913 1.20591 .05147 Post 442 5.1290 1.25053 .05948 Q 20: Sticking to Plan Pre 543 4.9871 1.12413 .04824 Post 441 4.9569 1.07045 .05097 Q 21A: Manager Sticks Pre 539 4.8720 1.26003 .05427 To Plan Post 434 4.9747 1.18177 .05673 Q 21B: Manager Sees Pre 539 5.2263 1.19041 .05127 New options Post 437 5.0870 1.31914 .06310 Q 21C: Quantitative Pre 535 4.5159 1.46214 .06321 Economical Decisions Post 430 4.5535 1.40948 .06797 Q 21D: Testing of New Pre 538 4.1004 1.30647 .05633 Work Methods Post 436 4.0963 1.35269 .06478 Q 21Manager: Seeks To Pre 531 3.7043 1.38631 .06016 Minimise Uncertainty Post 430 3.6674 1.32700 .06399 Q 21F: Uncertainty As Pre 523 3.9178 1.30242 .05695 Possibility Post 425 3.9976 1.18461 .05746 Q 21G: Independent Pre 535 4.4916 1.33219 .05760 Workgroups Post 429 4.4079 1.32165 .06381
9
Q 21H: Clear Distribution Pre 519 4.3526 1.36452 .05990 Of Responsibility Post 423 4.3522 1.33004 .06467 Q 21I: Boundary Pre 539 4.8887 1.42816 .06152 Crossing Meetings Post 435 4.7816 1.43383 .06875 Q 22A: Resistance Pre 543 4.1713 2.38037 .10215 Post 438 3.8973 2.36603 .11305 Q 22B: Ego Expansion Pre 537 3.9516 2.10399 .09079 Post 436 3.7959 2.14253 .10261 Q 22C: Action Prior Pre 535 3.3645 2.15240 .09306 To Analysis Post 439 3.2711 2.09954 .10021 Q 22D: Tactics Pre 540 3.1296 2.36272 .10168 Post 438 3.3105 2.32279 .11099 Q 22E: Avoidance Pre 537 4.8603 2.25291 .09722 Post 437 4.7025 2.31733 .11085 Climate Pre 554 4.8502 .82538 .03507 Post 444 4.8539 .83027 .03940 Efficacy Pre 552 4.8332 .69113 .02942 Post 444 4.8372 .68723 .03261 Creativity Pre 552 4.3191 .70506 .03001 Post 444 4.3346 .74348 .03528 Defensive Pre 548 3.9009 1.65854 .07085 Post 443 3.7997 1.63480 .07767
Figure 2. Pre–Post Poll
Note that Q q7, 18, and 19 were used in Post only
Correlations of Index–Post
The interpretation of each question, on the part of respondent as well as evaluator, contains the risk
of “error”; therefore factors and dimensions are often built of several questions (observations).
Statistical certainty of the emerging results and conclusions will that way be greater if index is used.
This material uses separate questions from existing index, but it is, despite that, necessary to
“remake” each index to obtain statistically valid conclusions.
Climate Index
The organisation climate of creativity and innovation instrument is made of ten dimensions with
five questions from each. We are using here one question a dimension, and consequently, our
Climate Index comprises one question per dimension, a total of ten questions.
Efficacy and Creativity Indexes
Based upon his research on experimenting, complexity, uncertainty, and outlook on planning and
efficacy, Professor Niclas Adler developed the query array we are using and the questions relate to
definitions of efficacy and creativity logics. Each query group forms each index.
Defensive Index
The instrument of offensive and defensive action comprises fifteen defensive variables that form
groups of five factors each. We are using one question per factor and the defensive index will make
up the defensive interaction pattern by using one question per factor, in all five questions.
10
Effect-1 and Effect-2 Indexes
These indexes were developed by Norrgren et al in their research on Change Strategies, and we are
using them in their original version.
Climate Effective Creative Defensive Effect-1 Effect-2
Climate Pearson Correlation
1 .369(**) .675(**) -.614(**) ,217(**) -.124(**)
Sig. (2-tailed) .000 .000 .000 .000 .010 N 444 444 444 443 436 432 Efficacy Pearson
Correlation .369(**) 1 .461(**) -.229(**) .113(*) -.121(*)
Sig. (2-talied) .000 .000 .000 .018 .012 N 444 444 444 441 436 432 Creativity Pearson
Correlation .675(**) .000 1 -.384(**) .297(**) -.116(*)
Sig. (2-tailed) .000 .000 .000 .000 .000 .016 N 444 444 444 443 436 432 Defensive Pearson
Correlation -.614(**) -.229(**) -.384(**) 1 -.100(*) .094
Sig. (2-tailed) .000 .000 .000 .036 .050 N 443 443 443 443 436 432 Effect-1 Pearson
Correlation .217(**) .113(*) .297(**) -.100 1 -.627(**)
Sig. (2-tailed) .000 .018 .000 .036 ,000 N 436 436 436 436 437 432 Effect-2 Pearson
Correlation -.124(**) -.121(*) -.116(*) .094 -.627(**) 1
Sig. (2-tailed) .00 .012 .016 .050 .000 N 432 432 432 432 432 433
** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05evel (2-tailed)
Figure 3. Correlations of Index–Post
Note: Index hyphenation comes from “reversed” scale compared to other indexes.
11
Climate Index
Below are the polls of each of thirty-six workplaces that comprise the study. According to the
Climate Index Table, Pre–Post changes generally are minor.
Workplace Climate
Pre Post Change
Std Std
Average Deviation Average Deviation (Post-Pre)
PR1 4.39315 1.090628 4.8 1.137927 0.406850395
PR2 4.128917 1.155459 4.281842 1.103937 0152924726
PR3 5.020174 1.115093 4.84207 0.97737 -0.178104026
PR4 4.240847 1.293271 4.075211 1.069594 -0.165635755
PR5 4.752632 1.230994 4.338739 1.135114 -0.413892512
PR6 4.383877 1.37143 4.136679 1.023521 -0.247197954
PR7 4.905 1.165611 4.642704 1.129312 -0.262295754
PR8 5.113618 0.928414 4.71707 1.08553 -0.396548018
PR9 4.273513 1.165967 4.369231 0.873227 0,095718187
PR10 3.945 1.256942 4.324265 1.221715 0.379264706
PR11 3.782172 0.895939 4.57 0.956798 0.787827663
PU1 4.870514 0.874079 4.743527 0.886806 -0.126987248
PU2 4.662382 1.06365 5.187864 1.062938 0.525482419
PU3 5.229166 0.741974 4.968494 0.949108 -0.260672431
PU4 4.522865 1.48285 4.81 1.138456 0.287134906
PU5 4.995411 1.234134 4.853556 1.179172 -0.141855422
PU6 4.944737 1.077758 4.571429 1.025272 -0.373307995
PU7 5.264595 1.007161 5.478655 0.815624 0.214060476
PU8 4.973415 0.900481 5.005882 1.132075 0.032467498
PU9 4.799277 0.906504 4.379338 1.044758 -0.419939217
PU10 5.39069 1.006554 5.40976 1.013739 0.019070213
PU11 4.169565 0.930356 4.255348 1.031992 0.08578261
PU12 4.945731 0.867987 4.73416 0.882713 -0.211570273
PU13 4.49476 1.37283 4.626041 1.375812 0.131281097
PU14 4.825118 0.830874 4.811023 1.177738 -0.014094434
PU15 5.406061 0.71206 5.4 0.993373 -0.006060606
PU16 4.425905 0.850319 4.457377 1.168921 0.031471178
PU17 5.221739 0.98667 4.489992 0.984209 -0.731747502
PU18 4.499617 1.005422 5.217155 0.890879 0.717537836
PU19 4.001487 1.279016 4.42759 0.931535 0.42610276
PU20 5.455 0.903649 5.304386 1.069702 -0.150614035
PU21 4.674964 1.170538 5.311438 1.022799 0.63647387
PU22 5.285714 0.9778 4.783333 0.969096 -0.502380952
PU23 4.701299 0.939515 4.696154 1.144728 -0.005144855
PU24 5.41042 0.926218 5.705 0.701171 0.294579832
PU25 3.480435 1.634027 3.857143 1.030273 0.376708075
Average PRIVATE Workplaces 4.448991 1.151795 4.463437 1.064913 0.014446514
Average PUBLIC Workplaces 4.826035 1.027297 4.859386 1.024916 0.033351112 Figure 4. Climate Index Average Values from Pre–Post Polls, All Workplaces.
Note that PR refers to private workplaces, and PU refers to public workplaces.
12
Efficacy Logics Index
Below are the polls of each of thirty-six workplaces included in the study. According to the
Efficacy Index Table, Pre–Post changes generally are minor.
Workplace Efficacy
Pre Post Change
Std Std
Average Deviation Average Deviation (Post-Pre)
PR1 4.590902 1.057071 4.828571 1.104401 0.237669904
PR2 4.518519 1.132043 4.434879 1.198681 -0.08363965
PR3 4.73892 1.356374 4.648166 1.039104 -0.09075376
PR4 4.483987 1.351157 4.277793 1.495893 -0.20619344
PR5 5.300752 1.222847 4.640507 0.931521 -0.66024492
PR6 4.86646 1.098992 4.318989 0.985525 -0.54747024
PR7 4.904135 1.023873 4.626053 1.120399 -0.27808217
PR8 4.738649 1.069989 4.598964 1.256155 -0.13968531
PR9 4.897708 1.189398 4.991592 1.137212 0.093884142
PR10 4.267126 1.298371 4.757878 1.107779 0.490752003
PR11 4.489708 1.12415 4.557143 0.93441 0.067435116
PU1 4.602286 0.870194 4.757368 0.745905 0.155081678
PU2 4.695174 0.962024 4.883228 1.113336 0.188054469
PU3 4.591417 0.878902 4.517259 0.930681 -0.07415806
PU4 4.414705 1.430307 5.1 1.250212 0.685295139
PU5 4.631154 1.339082 4.504861 1.262211 -0.12629278
PU6 4.470301 1.299934 4.252781 1.32343 -0.21751972
PU7 5.006616 1.111981 4.939421 1.102286 -0.0671956
PU8 4.828513 1.178661 5.285714 1.000077 0.457200822
PU9 4.844627 1.106093 4.457455 1.191227 -0.38717185
PU10 4.672659 1.484278 5.073056 1.151388 0.40039683
PU11 4.770186 1.025395 4.514459 1.175611 -0.25572692
PU12 4.266088 1.328545 4.288209 0.883448 0.022121031
PU13 4.748348 1.077512 4.942161 0.768067 0.193813023
PU14 4.154934 1.156731 4.440789 1.018695 0.28585474
PU15 4.961404 0.932557 4.775103 1.137445 -0.18630005
PU16 4.578818 0.772342 4.628773 0.916484 0.049954682
PU17 5.049689 0.891177 4.581518 1.086303 -0.46817192
PU18 4.382276 1.146133 4.580657 1.03108 0.198380982
PU19 4.701533 1.504206 4.66149 1.14066 -0.0400439
PU20 5.470301 1.220075 5.298835 1.296228 -0.17146543
PU21 4.552068 1.143138 5.108043 0.98837 0.555975612
PU22 4.489796 1.313265 4.285714 1.11971 -0.20408163
PU23 4.801793 1.056777 4.771868 1.029351 -0.02992531
PU24 5.26652 1.074686 5.327444 1.062769 0.060924098
PU25 4.301013 1.461546 4.397959 1.261338 0.096946559
Average PRIVATE Workplaces 4.708806 1.174933 4.607321 1.119189 -0.10148439
Average PUBLIC Workplaces 4.690089 1.150622 4.734967 1.079452 0.04487786 Figure 5. Efficacy Index Average Values from Pre-Post Polls, All Workplaces.
Note that PR refers to private workplaces, and PU refers to public workplaces.
13
Creativity Logics Index
Below are the polls of each of thirty-six workplaces that comprise the study. According to the table
Creativity Index changes of Pre–Post polls are in general minor; however, some questions display
significant changes.
Workplace Creative Logics
Pre Post Change
Std Std
Average Deviation Average Deviation (Post–Pre)
PR1 4.001553 1.30249 4.133333 1.313112 0.131779951
PR2 3.782368 1.242158 3.889766 1.287262 0.107398244
PR3 4.278713 1.346802 4.153407 1.119485 -0.125306274
PR4 3.635791 1.273988 3.643734 1.445613 0.007942995
PR5 4.37037 1.325273 4.038025 1.038887 -0.332345262
PR6 4.021739 1.209623 3.770064 1.069197 -0.251674927
PR7 4.268194 1.211279 4.312819 1.167822 0.044625422
PR8 4.566303 1.112897 4.184211 1.199217 -0.382092375
PR9 3.956878 1.096412 3.881766 1.039231 -0.075111701
PR10 3.637509 1.320548 4.026144 1.436264 0.388634713
PR11 3.735435 1.118527 4.077778 1.18477 0.342342318
PU1 4.240675 0.928908 4.360147 0.889914 0.119472691
PU2 3.95719 0.959047 4.588542 0.947736 0.631351785
PU3 4.394254 1.051055 3.95237 0.979687 -0.441883918
PU4 4.080371 1.267525 5.044444 1.123778 0.964073496
PU5 4.242111 1.117019 3.799066 1.420808 -0.443045104
PU6 4.638596 0.971058 4.165578 1.336338 -0.473018035
PU7 4.566467 1.11628 4.901488 1.172396 0.335021241
PU8 4.204344 0.902482 4.474673 1.075171 0.270329197
PU9 4.491863 1.165024 4.105987 1.304227 -0.385875518
PU10 4.294484 1.095251 4.551539 1.153333 0.257055031
PU11 3.826087 1.062485 3.957574 1.004737 0.131486694
PU12 4.342983 1.057933 4.319253 0.865766 -0.023729822
PU13 3.88824 1.197771 4.243004 1.166129 0.354764087
PU14 4.2838 0.909186 4.338208 1.040706 0.054408334
PU15 4.62381 0.915199 4.675444 0.904459 0.051633987
PU16 3.961437 0.984234 4.146532 1.360911 0.185094588
PU17 4.676329 1.092369 4.09077 1.10181 -0.585558544
PU18 4.014339 1.024935 4.553694 0.843506 0.539354975
PU19 3.818483 1.360789 3.818813 1.155665 0.000330513
PU20 5.072807 1.110426 4.948506 0.97479 -0.124301494
PU21 4.295668 1.225784 4.687259 1.0638 0.391591099
PU22 4.174603 1.016178 4.092593 0.959195 -0.082010582
PU23 4.116162 1.079242 4.051111 1.019762 -0.065050505
PU24 4.807362 0.938435 4.949708 0.93253 0.142345466
PU25 3,450373 1,61506 3,52381 1,203922 0,073436748
Average PRIVATE Workplaces
4.023169 1.232727 4.010095 1.209169 -0.013073354
Average PUBLIC Workplaces 4.258513 1.086547 4.333605 1.080043 0.075091056
Figure 6. Creativity Index Average Values from Pre-Post Polls, All Workplaces.
Note that PR refers to private workplaces, and PU refers to public workplaces.
14
Two questions from the Creativity Index, i e number 13 and 14, show significant improvement
subsequently to the AIRIS run:
1. At the post-poll one (more) often meets new people and get new perspectives on one’s work
(t=2.43, p<.05). Question 13
2. At the post-poll one (more) often breaks with conventional patterns (t=2.36, p<.05).
Question 14
Another question from Creativity Index shows a tendency towards ensured improvement:
3. At the post-poll there is a tendency for describing as a good one a boss that sees new
possibility and adapts the business accordingly (p<.10). Question 21b
Defensive Action Patterns Index
Below are the polls of each of thirty-six workplaces that comprise the study. According to the table,
changes in the Defensive Index of Pre–Post generally are minor; however, one question shows a
significant change.
Workplace Defensive Action
Pre Post Change
Std Std
Average Deviation Average Deviation (Post-Pre)
PR1 4.296002 2.118531 4.026667 2.169609 -0.269335394
PR2 4.234667 2.204394 4.010526 2.21679 -0.224140351
PR3 2.940312 1.765412 2.680406 1.839103 -0.259905907
PR4 5.33395 2.196699 4.555861 1.867911 -0.778089816
PR5 4.357895 2.310711 4.14285 1.757893 -0.215044283
PR6 5.058333 2.620673 4.428625 2.488585 -0.629708538
PR7 4.28 2.534579 4.258057 2.129718 -0.021942574
PR8 3.929149 1.971385 4.694737 2.020509 0.765587914
PR9 4.618383 2.655336 4.284615 1.947933 -0.33376722
PR10 5.665789 2.08476 4.364706 2.227642 -1.301083591
PR11 5.018097 2.219563 3.871111 2.234414 -1.146986
PU1 3.246724 1.653651 3.467631 2.097065 0.220906586
PU2 2.682868 1.940084 2.37497 1.846532 -0.307898458
PU3 2.801408 1.892877 3.034063 2.037672 0.232654374
PU4 3.865694 2.062016 4.22 2.450099 0.354306361
PU5 2.648243 1.667564 3.147677 1.808902 0.499433528
PU6 3.88 1.972291 4.111189 2.242533 0.231188563
PU7 3.76888 2.369079 2.734591 1.432418 -1.034288872
PU8 2.936986 1.788536 3.270588 2.17469 0.333602251
PU9 3.246574 1.683065 4.212091 2.091025 0.965516695
PU10 2.386973 1.48027 2.322356 1.253815 -0.064616846
PU11 5.12253 2.005874 4.532737 2.082715 -0.589792482
PU12 3.292637 2.038849 4.010554 1.732697 0.717916589
PU13 4.149782 2.589678 3.376132 2.639888 -0.773649799
PU14 3.455414 1.797329 4.031536 2.302318 0.576122665
PU15 2.761905 1.490958 3.666667 2.364063 0.904761905
PU16 4.140223 1.685019 3.381665 2.885474 -0.758557438
PU17 3.504348 1.868771 3.999617 1.926188 0.495269569
PU18 3.933577 2.212051 2.437991 1.723623 -1.49558561
PU19 3.875446 2.174722 3.676388 1.965912 -0.19905855
PU20 4.1 2.308727 3.531871 2.22285 -0.568128655
PU21 3.296401 2.280376 2.950101 2.544558 -0.346300549
PU22 2.885714 1.596425 3.733333 2.443077 0.847619048
PU23 3.581385 1.852097 3.320308 2.032204 -0.261077589
PU24 3.022584 1.802141 2.607368 1.514462 -0.415215613
15
PU25 5.239526 2.576708 4.542857 2.543795 -0.696668549
Average PRIVATE Workplaces 4.521143 2.243822 4.119833 2.081828 -0.401310524
Average PUBLIC Workplaces 3.513033 1.951566 3.467771 2.094343 -0.045261635
Figure 7. Average Defensive Index Values from Pre–Post Polls, All Workplaces.
Note that PR refers to private workplaces, and PU refers to public workplaces.
One question from Defensive Index, resistance (number 22a in our poll), shows significant
improvement subsequently to the AIRIS run:
1. There is a tendency toward reduced resistance during post-poll (p<.10). Question 22a
Effect-1 and Effect-2 Indexes
Workplace Effect-1 Effect-1
Post Post
Std Std Average Deviation Average Deviation
PR1 2.175137 0.667275053 2.641026 0.49770794
PR2 1.960197 0.625698699 2.900877 0.39379674
PR3 2.142196 0.898719049 2.433595 0.58756529
PR4 1.644375 0.704797502 2.697962 0.36920841
PR5 1.438868 0.770471241 2.769791 0.69044197
PR6 1.732039 0.655465434 2.770141 0.52243935
PR7 1.43693 0.557908077 2.900939 0.35572469
PR8 2.059211 0.742484685 2.736842 0.45902241
PR9 2.409455 0.665002832 2.411616 0.41534547
PR10 1.742188 0.711134905 2.770833 0.45222601
PR11 2.120536 0.64010185 2.619048 0.48233648
PU1 2.094573 0.794021375 2.546396 0.53327426
PU2 2.002519 0.794376115 2.497631 0.43028123
PU3 1.635193 0.560816552 2.615863 0.2580973
PU4 2.143056 0.643342786 2.566667 0.48836036
PU5 1.802268 0.805397943 2.53009 0.64671285
PU6 2.010649 0.76888288 2.609936 0.74366457
PU7 2.440022 0.678332813 2.350502 0.49228712
PU8 1.878156 0.797094155 2.973389 0.66334468
PU9 1.98401 0.757041335 2.529087 0.49205724
PU10 1.630658 0.787170481 2.536794 0.42056449
PU11 1.303796 0.481386435 2.746783 0.68747071
PU12 1.760458 0.585434055 2.742939 0.17213259
PU13 1.855497 0.632654386 2.643526 0.22222222
PU14 1.812174 0.757826968 2.65095 0.53944611
PU15 2.158208 0.736984382 2.592105 0.42688479
PU16 2.102595 1.025745754 2.499769 0.69716312
PU17 2.050175 0.693205689 2.397705 0.47787926
PU18 2.405918 0.704012556 2.425473 0.49132385
PU19 1.974988 0.663974242 2.641997 0.54050248
PU20 2.194079 0.731368772 2.581546 0.76183089
PU21 2.145821 1.012062793 2.551663 0.57735027
PU22 2.0625 0.737733013 2.666667 0.82724922
PU23 1.226852 0.522909363 3.098765 0.4078334
PU24 2.23125 0.560167339 2.7 0.33325942
PU25 2.134524 0.607571353 2.809524 0.41462633
Average PRIVATE Workplaces 1.896466 0.694459939 2.695697 0.47507407
Average PUBLIC Workplaces 1.961598 0.713580541 2.620231 0.50983275
Figure 8. Average Effect-1 and Effect-2 Indexes Values from Pre–Post Polls, All Workplaces.
Note that PR refers to private workplaces, and PU refers to public workplaces.
16
Although the Index level does not show great effects, we can do some interesting observations at
single question level. Eight out of ten respondents think that AIRIS contributed to working climate
as well as co-operation improvement in the company (four out of ten say AIRIS contributed to a
large extent). Four out of ten say AIRIS contributed to productivity and quality enhancement.
These results have not been validated statistically, although the work climate and co-operation
variables concord well with interviews and other evaluations (see qualitative results below).
17
Conclusions
Assembled qualitative data was statistically processed in order to identify significant changes and
correlations. There are two significant pre–post diversities (p<.05), and there are two tendencies
toward difference (p<.10).
1. At post-poll one (more) often meets new people and get new perspectives on one’s work
(t=2.43, p<.05). Question 13
2. At post-poll one (more) often breaks with conventional patterns (t=2.36, p<.05) Question 14
3. At post-poll there is a tendency to describe as a good one a boss who sees new possibility
and adapts the business accordingly (p<.10). Question 21b
4. There is a tendency toward reduced resistance at post-poll (p<.10). Question 22a
In addition, we can conclude that four out of five respondents believe that AIRIS made a positive
contribution to the working climate and co-operation at the workplace and more than a third of co-
workers say AIRIS contributed to medium or high degree of qualities such as:
Improved cooperation
Improved working climate
Enhanced quality
Enhanced productivity
In general, the results indicate reduced resistance against change and space expansion for creative
logics. These workplaces demonstrate better capabilities of managing unpredictability and
alternative perspectives. More comfort and improved cooperation at the company are other common
effects. We may also note that the effects during the project term seem to be greatest at the
individual level.
18
Sick Leave Results
Some workplaces that previously participated in AIRIS hold that the result of the participation was
lower sick leave levels. Since 2006 we have tried to obtain relevant information from the
workplaces; however, obtaining comparable figures over larger time spans proved difficult,
probably depending upon a large number of non-verifiable factors. But the AIRIS sister project
Genklang offered relatively favorable conditions. All units involved are part of Vara Municipality
and are generally using a common account system. And in addition, the other non-participating
municipal units exist as a control group.
Totally, nineteen workplaces within the Vara Municipality each performed an AIRIS run over a
span of two years. Therefore, we can compare changes of sick leaves between those units that
participated in Genklang with all other units within the Vara Municipality. The changes refer to
accumulated sick leaves from August 2006 to July 2007 compared to the period between August
2005 and July 2006, containing changes of short term leaves (one to fourteen days) as well as long
term leaves (sixty days or more). Registered changes thus refer to the period of the Genklang run
with the base value of the nearest preceding twelve month period (=100).
Units Short Term Sick Leaves Long Term Sick Leaves
Genklang 1 150.0 100.0
Genklang 2 190.0 62.6
Genklang 3 51.3 35.9
Genklang 4 127.0 58.8
Genklang 5 99.7 301.9
Genklang 6 81.5 75.5
Genklang (unit 1-6) 106.0 62.9
Vara Municipality 118.7 105.7 Figure 9. Changes of short and long term sick leaves for the duration of the Genklang project (year 1)
compared to the previous twelve month period
The statistics above show that the increase of short term sick leaves in units that fully participated in
the Genklang Vara run were just a third of the average municipal increase. The Genklang units
show considerable improvement of long term sick leaves while the municipality as a whole shows
some degree of decline. We can also note that short term sick leaves declined in as many units as it
improved while long term sick leaves improved in all units but one.
19
We see the development of the Genklang Vara units year two as follows:
Units Short Term Sick Leaves Long Term Sick Leaves
A -12 -71
B -24 -71
C -9 602
D -16 131
E 170 Nil previous year
F -22 -100
G 48 34
H -7 Nil previous year
Genklang
second year
(A–H) -8,9 16,6
Vara
Municipality -6,5 -3,6 Figure 10. Percent change of short and long term sick leaves the Genklang year two
compared to the previous twelve months period.
Most workplaces that participated in the Genklang Vara project the second year also showed
improved short term sick leave development compared to that of all municipal workplaces during
the same period. Although the difference is not as large that of Genklang Vara the first year, it is
clearer, and its value is estimated to about SEK 600,000 for the Vara Municipality. On the other
hand, the changes of long term sick leaves indicate that they are probably due to other factors.
We may also look at the development of workplaces that participated in the second Genklang Vara
project previous to project launch, i e, parallel to the run of the first Genklang Vara. We may,
correspondingly, also look at the sick leave development of the workplaces that participated in the
first Genklang Vara project the year second to participation, i e, parallel to the run of the second
Genklang Vara. Would this positive development prevail?
Short term sick leave
development the year prior to
the Genklang Vara project
Short term sick leave
development the year sub-
sequent to the Genklang Vara
project
Genklang Vara work-
places first year
1,91
Genklang Vara Work-
places second year 21,82
Vara Municipality all
workplaces 18,7 -6,5 Figure 11. Percent change of short term sick leaves for each year of the Genklang Vara project (six and
eight workplaces respectively) in comparison with the total of Vara Municipality.
Here we see that the units, that through Genklang Vara second year showed more positive
development than did Vara Municipality totally, also were worse off than Vara Municipality the
year prior to the project. Moreover, we see that the workplaces, that participated in the first year,
and then showed better development than Vara Municipality totally, were significantly worse off
the year subsequent to the project, compared to Vara Municipality. The workplaces of Genklang
Vara did not manage to maintain its positive development. Possible causes for this may be:
20
- None of the workplaces managed to continue on/keep up with the line of action chosen
(complete change is not achieved in one project year only but takes longer, and the relay
race stick was handed over to the management/organization to take the process further).
- This development is an instance of the so called Hawthorne Effect (the everyday change that
a project like Genklang Vara implies, with internal and external attention to the participating
workplaces, gives a very positive impact that consequently vanishes when the project is
terminated or becomes “normal”)
Conclusions
Changes of sick leave indicate that their short term decreased to a greater extent compared to Vara
Municipality totally during the year of the AIRIS run. However, the change did not seem to leave
permanent results subsequently to the project termination. Conclusions are uncertain, partly due to a
limited number of observations, and also because sick leaves are probably subject of many causal
factors.
21
Qualitative Results
Phone interviews were performed with representatives of half the number of participating
workplaces (about twelve) in AIRIS 2003–2006. We have consulted all participating workplaces;
however, our selection was determined by those who responded. Every workplace we have
contacted said they wanted to cooperate in the study. Workplaces that were involved in the Pilot
Project are, however, not part of this study.
The interviews featured questions about what one wanted to obtain by the project and the results
achieved, what happened in the company after project termination, how the results are maintained,
what position one takes today, and the effects AIRIS caused.
What Did the Participation Achieve?
Today, all workplaces we interviewed are very pleased with AIRIS. The objectives one say one
entertained are consistent with the results one say one obtained. Many of the workplaces say that
AIRIS had the following consequences:
AIRIS united a group of coworkers or organised units, geographically or
structurally separated. The project brought a sense of companionship and
solidarity to the group.
“AIRIS became an arena for encounters across boundaries, with
social connections, within as well as without the workplace. We
arrived at greater mutual understanding, and AIRIS meant a
huge exchange of competence.”
AIRIS caused enhanced communication, within an organisation unit as well as
between organization units.
AIRIS caused improved workplace climate.
AIRIS caused improved self-image among co-workers. Managers achieved
knowledge about people and about their co-workers through the AIRIS
project.
“The Project made people grow. They coped better with the
taxing challenges we faced during the years that followed.”
AIRIS caused co-workers to be more courageous and more open to new stuff,
leading to greater conceptual “space” and a more open atmosphere.
“If you do what you’ve always done you’ll get what you’ve
always got. You have to look beyond the predictable.”
“Following norms seldom lead to the high-quality solutions
that would generate innovation. Instead, it is what you dare
to do square to what is expected.”
AIRIS caused workplaces to adopt a broader perspective on their own
business. AIRIS made the staff begin to challenge structures and patterns
that proved less meaningful.
22
”AIRIS shakes your normal everyday routines; you’ll have
to step out of your own problems and view matters in a new
way.”
Some of the workplaces talk about the implications of the project for the staff as a kind of mental
maintenance; for example, the pros and cons of the nonexistent demand for “concrete results” in the
project. This characteristic of the project; getting into something so open and liberating and being
allowed to seize the moment, was described as a challenge both for the individual and the
organisation.
Other long term effects are more difficult to captivate. This is partly because AIRIS has often been
a component in a development process parallel or sequential to other projects, and partly because
the workplaces have undergone major and minor changes subsequently to AIRIS that it is today
hard to say something other than that most people agree that AIRIS made a positive contribution to
the development that did happen.
From Skådebanan we received the estimated PR value of some of the workplaces, and one value
assessment of the effects that the AIRIS project left them:
PR Value according to Skådebanan’s Assessment
SCA, Lilla Edet 4 minutes Swedish National Televised News 2007/04/11,
corresponding to app SEK 6.5 million (estimate based
on the cost of TV commercials at TV4 and Dagens
Industri’s estimates of editorial material).
AstraZeneca, Mölndal 17 press clips, corresponding value app SEK 8
million (based on Dagens Industri’s estimates of
editorial material)
ABB Kabeldon, Alingsås 20 minutes at Swedish National Television Kobra
corresponding to about SEK 10 Million; Goteborg’s
Culture Party in a major paper Göteborgs-Posten,
corresponding to SEK 0.5 Million. Other press clips
in magazines corresponding to SEK 1 Million; in
total about SEK 11.5 Million.
Housing Company Bostadsbolaget’s Estimations
A SEK 200,000 AIRIS run produced the following yield distributed onto about 80 individuals:
• Reduced health expenses about SEK 1.3 million distributed over 2 years.
• The Satisfied Customer Index raised 10 % and Bostadsbolaget now topping the group.
• The Satisfied Co-worker Index raised 16 %.
• Media value estimated to about SEK 1.5 million.
Total value for Bostadsbolaget: about SEK 5 million.
23
Conclusions
Significant trends of the Quantitative Study speak of “seeing new people and getting broader
perspectives on work”, and “breaking conventional patterns”. These phrases may be viewed as
corresponding to characteristics often ascribed to artists, that is, the capability of working with
alternative views and working with the unpredictable. The trends of the quantitative material also
speak of “reduced resistance to change”, and “a good boss sees new possibility and adapts his
business accordingly”. Such trends signal increased inclination to change, openness towards new
things, both directly (less resistance) and indirectly (seizing new possibility). The supplementary
qualitative material also repeats the experience, at various levels in the participating organisations,
of increased cooperation and solidarity as well as a commonly enhanced working climate as a result
of the AIRIS project.
Figure 12. How a workplace is affected by an intervention based on art and culture
Artist
Capability to Work With
the Unpredictable
Capability to Work With
Alternative Perspectives
Enhanced Cooperation
and Solidarity
Improved
Working Climate
Reduced Resistance to
Change
Increased
Inclination to
Change
24
Here we may draw parallels to the discussion on Comfort Zones we have seen from other studies.
There is certain reluctance among executives and co-workers to leave their comfort zone to test new
approaches, or venturing to challenge dominant suppositions that govern decisions and actions.
Figure 13. Managers and co-workers’ imprisonment in Comfort Zones (from Adler & Beer)
By the AIRIS project and the artists we could say that managers and co-workers at the participating
workplaces received an impulse that forced them out of their comfort zones, or, in several cases,
broadened them, possibly making larger space for change and innovation. According to Elmquist
(2007), the processes of product development within large organisations are designed mainly for
developing well defined products in an efficient way, and that innovative concepts have difficulties
in impacting these processes. That leads to two problems; one, the assumption that the actors in the
process possess enough knowledge to respond to the right signals; and two, that the organisations
already have identified the opportunities and therefore are in a position to select among them. In
order to develop innovation processes organisations need first, enhance their capability to absorb
new knowledge through recognizing and understanding potentially valuable new knowledge, and
second, develop new concepts and commercial innovations by actively experimenting and creating
potentially valuable new knowledge. The integration between generative and absorptive processes
is considered a fundamental dimension to enable both an efficient NPD (New Product Develop-
ment) process and the development of more innovative products (ibid).
Percieved to support and
solidify existing
dominant assumptions
Percieved to disturb or
threaten existing
dominant assumptions
Percieved to
support and solidify
contemporary work
procedures
Percieved to disturb
and counteract
contemporary work
procedures
Comfort Zone
25
References
Adler, N. (1999), Managing Complexity in Product Development – Three Approaches. EFI Stockholm School of Economics.
Beer, M., Eisenstat, R. A. & Spector, B. (1990), The Critical Path to Corporate Renewal. Harvard Business School Press
Beer, M. & Nohria, N. (2000), Breaking the Code of Change. Harvard Business School Press, Boston MA
Ekvall, G. (1996), Organizational Climate for Creativity and Innovation European Journal of Work and Organizational Psychology 5(1): 105-123
Elmquist, Maria (2007) Enabling Innovation – Exploring the Prerequisites for Innovative Concepts in R&D, Chalmers tekniska högskola (Doctoral Thesis)
Eriksson, M. & Sundgren, M. (2005), “Managing Change: Strategy or Serendipity – Reflections from the Merger of Astra and Zeneca”, Journal of Change Management, Vol 5 No 1: 15-28.
Hatchuel, A. (2001), “The Two Pillars of New Management Research” British Journal of Management 12 (Special Issue): S33-S39.
Hatchuel, A., Le Masson, P. & Weil, B. (2001). From R&D to R-I-D: Design Strategies and the Management of "Innovation Fields". Paper presented at the EIASM 8
th International Product Development Management
Conference, Enschede, the Netherlands.
Kylén, S. (1999). Interaktionsmönster i arbetsgrupper - offensiva och defensiva handlingsrutiner. Department of Psychology Göteborgs Universitet & FENIX Research Programme (Doctoral Thesis)
Norrgren, F., Hart, H. & Schaller, J. (1996), Förändringsstrategiers effektivitet, CORE WP1996:3
Styhre, A. & Eriksson, M. (2007) Bring in the Art and Get the Creativity for Free – A Study of the Artists In Residence Project, Creativity and Innovation Management
Sundgren, M. (2004). New thinking, Management Control & Instrumental Rationality - Managing Organizational Creativity in Pharmaceutical R&D. Gothenburg: Institute for Management of Innovation and Technology.