University of Southern California Dissertations and Theses

Influencing individual innovation through technology features that support cross -departmental understanding

(USC Thesis Other)

Influencing individual innovation through technology features that support cross -departmental understanding

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Influencing Individual Innovation Through Technology
Features that Support Cross-departmental Understanding
Copyright 2004
by
Ixchel Marika Faniel
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements of the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)
December 2004
Ixchel Marika Faniel
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
UMI Number: 3155406
Copyright 2004 by
Faniel, Ixchel Marika
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3155406
Copyright 2005 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
DEDICATION
To my ancestors who graciously raised me on their backs so I could get a
better view.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
ACKNOWLEDGEMENTS
I want to start by thanking my committee members, who have been patient
and supportive, as I have waded through this dissertation process. First, I am
grateful to Dr. Ann Majchrzak, my committee chair, not only for the quantity of time
that she has given me, but also for the quality of that time. I also thank Dr. Gareth
James, for his time and patience when helping me with statistics and Dr. Robert
Josefek, for continued encouragement and advice. Thank you to Dr. Janet Fulk and
Dr. Richard Clark, for the opportunity to spend time in their classroom. Taking their
classes helped shape my research interests.
Thank you to Dr. Omar El Sawy for encouraging me to apply to the doctoral
program at Marshall and continuing to provide support thereafter. I also thank Dr.
Arvind Bhambri and Dr. Valerie Folkes for their encouragement and support in the
early stages. Thank you to Deborah, Tina, and Elizabeth. They are simply the best.
Anything I wanted or needed, they did not hesitate to get it for me. I thank Richard
Bergin and Dr. Ram Chellappa for giving me the opportunity to see how teaching
should be done. I will try to keep up. Two other patient statistics professors I must
thank are Dr. Delores Conway and Dr. Bert Steece. I know I went over the office
hour limit. I also want to thank Nancy Stoffer, who didn’t mind me just dropping by
her office with questions. To IS students past and present, especially Ricky, who
helped me through the first year and Raymond, Shaosong, Jenny, and Dave who
were enthusiastic about sharing research ideas during lunchtime seminars. I also
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
iv
want to thank Ricardo, who reminded me not to take things so seriously and Marlene
who always called to see how I was doing.
Thank you to the KPMG Foundation, the PhD Project, and ISDSA for
providing continued emotional and financial support. I am especially grateful to
Bemie Milano and Tara Perino, who work tirelessly for their Ph.D. students. Special
thanks must also go to Stew Sutton and Kevin Kreitman. Their generosity and
support helped make this happen.
I am also grateful and have much love for my family and friends. I have too
many to name individually, but would like to mention a few. To my mom and dad,
how they managed to raise and educate five children is beyond me. I am truly
appreciative of their love, hard work, and dedication in guiding their children
through life. I also want to thank my brothers and sisters, Lance, Erik, Rhea, and
Roxanne. I could not have asked for better siblings, they have constantly
encouraged, supported, loved, and protected me, as well as made me laugh until my
belly hurt. I also thank Ayanna and Noah for making me smile. Lastly, I am
grateful to Uncle Tommy, Bea, Karen and Holly for loving and supporting me from
birth as if I were their own. Special thanks to Karen, words cannot express my
gratitude.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
V
TABLE OF CONTENTS
DEDIC ATIO N.................................................................................................................................. ii
ACKNOW LEDGEM ENTS.........................................................................................................iii
LIST OF TABLES A N D FIGURES.......................................................................................viii
ABSTR AC T....................................................................................................................................... x
CHAPTER 1: EXECUTIVE SU M M A R Y ............................................................................... 1
1.1 M otivation........................................................................................................................... 2
1.2 Overview of Conceptual M odel and H y po th eses............................................ 3
1.3 Overview of Research Methodology..................................................................... 5
1.4 Sum m ary of Fin d in g s.......................................................................................................7
1.5 Sum m ary of Co n tribu tio n s..........................................................................................7
1.6 R oadmap to the D isser ta tio n...................................................................................10
CHAPTER 2: INTRO DUCTIO N..............................................................................................11
2.1 In no vatio n.......................................................................................................................... 12
2.2 Perceived V alue of Cross-departmental Knowledge on In n o va tio n. 14
2.3 Influencing Fa cto r s......................................................................................................15
2.4 Technology Features That B ridge Thought W orld B arr iers................17
2.5 Su m m a r y ..............................................................................................................................20
CHAPTER 3: CONCEPTUAL M ODEL................................................................................ 22
3.1 Theory of Thought W o r l d s....................................................................................23
3.1.1 Funds o f K nowledge.................................................................................................. 24
3.1.2 Systems o f M eaning................................................................................................... 25
3.1.3 Readiness fo r D irected P erception .......................................................................25
3.2. A ccessing Knowledge That Bridges Thought W orld B arriers 26
3.2.1 Knowledge Should Be A ssociated With the Expert Source............................27
3.2.2 Knowledge Should be Interpretable..................................................................... 27
3.2.3 Knowledge Should Be A daptable.......................................................................... 28
3.3 Cultural Support That B ridges Thought W orld B a r r ier s..................... 29
3.4 Theory of Hermeneutic In q u ir y ..............................................................................30
3.5 Technology Features that Support Cross-departmental
U n d e r st a n d in g................................................................................................................ 32
3.5.1 Supporting Associations Between Knowledge and the Expert S o u rce 33
3.5.2 Supporting Interpretable K now ledge...................................................................34
3.5.3 Supporting Adaptable Knowledge.........................................................................35
3.5.4 An Example o f Technology Features that Support Cross-departm ental
Understanding............................................................................................................... 36
3.5.5 Summary........................................................................................................................38
3.6 Control V a r ia b l e s........................................................................................................ 38
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
v i
3.6.1 P erceived Openness o f C om m unication............................................................. 39
3.6.2 P erceived Task Com plexity..................................................................................... 39
3.6.3 People-to-people Interaction................................................................................. 40
3.6.4 P erceived Ease o f Use o f Technology..................................................................41
3.1 C o n c e p tu a l M o d e l......................................................................................................... 42
3.8 Exploratory Qu est io n.................................................................................................43
3.8.1 P erceived Openness o f C om m unication............................................................. 44
3.8.2 People-to-people Interaction................................................................................. 45
3.8.3 P erceived Task Com plexity.....................................................................................45
3.8.4 P erceived Ease o f Use o f Technology..................................................................46
3 .9 Su m m a r y ............................................................................................................................ 47
CHAPTER 4: RESEARCH M ETH O D S................................................................................ 48
4.1 Research Se t t in g............................................................................................................48
4.2 Preliminary D ata Co llectio n................................................................................. 49
4.2.1 Participant Observation o f K M Committee M eetings..................................... 50
4.2.2 Structured Interviews with M anagem ent............................................................51
4.2.3 Sem i-structured Case Studies................................................................................. 54
4.2.4 Summary........................................................................................................................58
4.3 Survey M ethodology................................................................................................... 59
4.3.1 Survey Adm inistration...............................................................................................59
4.3.1.1 Two-W ave S u rvey............................................................................................ 60
4.3.1.2 Power A n alysis..................................................................................................61
4.3.1.3 A ssessing Non-response B ia s ........................................................................62
4.3.1.4 Sum m ary..............................................................................................................64
4.3.2 Survey Instrum ents.....................................................................................................64
4.3.2.1 U se o f Technology Features that Support Cross-departmental
Understanding.................................................................................................... 65
4.3.2.2 Perceived Openness of Com m unication.....................................................68
4.3.2.3 People-to-people Interaction.......................................................................... 69
4.3.2.4 Perceived Task C om plexity........................................................................... 71
4.3.2.5 Perceived Ease of U se o f T ech n ology........................................................ 72
4.3.2.6 Perceived Value of Cross-departmental Knowledge on Innovation..73
4.4 D ata A nalytic Tech niques........................................................................................ 77
CHAPTER 5: R E SU L TS.............................................................................................................79
5.1 First W ave D ata A nalysis..........................................................................................79
5.1.1 First Wave Tests fo r Discrim inant V alidity....................................................... 80
5.1.2 First Wave D escriptive Statistics and Correlations.........................................85
5.1.3 Examination o f F irst Wave D ata P rior to H ierarchical Regression
A n alysis........................................................................................................................... 86
5.1.4 F irst Wave H ierarchical Regression A n alysis.................................................. 88
5.1.5 Two M ajor Limitations o f First Wave D ata A n alysis..................................... 91
5.2 Second W ave D ata A n a l y s is....................................................................................93
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
v ii
5.2.1 Second Wave Tests for Discriminant Validity............................................. 93
5.2.2 Second Wave Descriptive Statistics and Correlations................................96
5.2.3 Examination of Second Wave Data Prior to Hierarchical Regression ....98
5.2.4 Second Wave Hierarchical Regression Analysis........................................99
5.2.5 Identification of Influential Second Wave Data Points............................ 101
5.2.6 Revisiting Hierarchical Regression Analysis with Influential Points
Removed........................................................................................................103
5.3 Hierarchical Regression A nalysis for the Exploratory Question .. 106
5.4 Su m m a r y ............................................................................................................................108
CHAPTER 6: DISCUSSION...........................................................................................110
6.1 Lim itations.......................................................................................................................110
6.1.1 Data Collection Procedures Took Place at One Site................................110
6.1.2 Two-wave Survey Methodology Reduced the Response Rate..................I l l
6.1.3 Two Survey Instruments Were Newly Developed......................................I l l
6.2 C o n tr ib u tio n s t o T h e o r y .......................................................................................... 112
6.3 Future Re se a r c h...........................................................................................................115
6.4 Contributions to Pr a c tic e.......................................................................................119
6.6 Co n clusio n.......................................................................................................................120
REFERENCES.................................................................................................................. 121
APPENDIX A ....................................................................................................................127
APPENDIX B ....................................................................................................................131
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
LIST OF TABLES AND FIGURES
Table 3.1. Technology Design Principles........................................................................33
Figure 3.2 Conceptual Model............................................................................................ 42
Table 4.1 KM Technologies Available for Use at Company A ....................................53
Table 4.2 Case Study Findings that Verify How Company A Employees Bridge
Cross-departmental B arriers.....................................................................................56
Table 4.3 Case Study Findings that Confirm Technology Features Exist to Help
Bridge Cross-departmental Barriers.........................................................................57
Table 4.4 Case Study Findings that Confirm Other Factors in the Work
Environment Should be Controlled..........................................................................58
Table 4.5 First Wave Characteristics for Response Groups and Total Sam ple..........62
Table 4.6 Second Wave Characteristics for Response Groups and Total Sample..... 62
Table 4.7 First Wave Variable Means over Three Weeks of Response ...............63
Table 4.8 First Wave Variable Means for Respondents and Non-respondents
of the Second W ave...................................................................................................64
Table 4.9 Items for IT Systems Available for Use at Company A .............................. 65
Table 4.10 Items and Factor Analyses for the Use of Technology Features that
Support Cross-departmental Understanding...........................................................66
Table 4.11 A Comparison of Items and Factor Analyses for Perceived Openness
of Communication: O ’Reilly and Roberts vs. Dissertation Study.......................69
Table 4.12 Items and Factor Analyses for People-to-people Interaction.................... 70
Table 4.13 Items and Factor Analyses for Perceived Task Complexity......................72
Table 4.14 A Comparison of Items and Factor Analyses for Perceived
Ease of Use of Technology: Davis vs. Dissertation Study...................................73
Table 4.15 A Comparison of Items and Factor Analyses for Perceived Value of
Cross-departmental Knowledge on Innovation: Jabri vs. Dissertation Study.... 75
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.1 First Wave Factor Solution with Varimax Rotation..................................... 81
Table 5.2 First Wave Descriptive Statistics and Correlations...................................... 84
Table 5.3 First Wave Hierarchical Regression Results..................................................90
Table 5.4 Second Wave Factor Solution with Varimax Rotation.................................93
Table 5.5 Second Wave Descriptive Statistics and Correlations..................................97
Table 5.6 Second Wave Hierarchical Regression Results...........................................100
Table 5.7 Second Wave Influential Observations.........................................................102
Table 5.8 Second Wave Hierarchical Regression Results Without
Influential Points...................................................................................................... 105
Table 5.9 Second Wave Exploratory Hierarchical Regression for Perceived
Value of Cross-departmental Knowledge on Divergent Thinking...................107
Table 5.10 Second Wave Exploratory Hierarchical Regression for Perceived
Value of Cross-departmental Knowledge on Convergent Thinking................ 108
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
ABSTRACT
In today’s world, organizations that are able to maintain competitive
advantage are those that can figure out ways to influence employee innovation.
Enabling employees to access knowledge from different areas of expertise within the
organization offers one alternative. This often means acquiring knowledge from
colleagues in other departments. However, organizations face a major challenge in
figuring out how to bridge the barriers to cross-departmental knowledge access,
barriers that lower individuals’ perceived value of cross-departmental knowledge on
innovation. This dissertation study couples the theory of thought worlds with the
theory of hermeneutic inquiry to determine whether the use of technology features
that support cross-departmental understanding can help. The proposed model
suggests the use of technology features that support cross-departmental
understanding is positively related to individuals’ perceived value of cross-
departmental knowledge on innovation. In the context of this dissertation study,
individuals’ perceived value of cross-departmental knowledge on innovation refers
to their perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking, two recognized components of innovation. The research began
with interviews and case studies with scientists and engineers in an aerospace firm.
This was followed by a two-wave survey of 874 members of the firm. Results
supported the proposed model. Individuals who used technology features that
support cross-departmental understanding believed that cross-departmental
knowledge helped their divergent thinking and convergent thinking. Moreover,
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
these relationships remained significant after controlling for other factors in the work
environment including: perceived openness of communication, people-to-people
interaction, perceived task complexity, and perceived ease of use of technology. The
findings from this dissertation study showed that coupling the theory of thought
worlds with the theory of hermeneutic inquiry provides a useful theoretical base for
designing knowledge management technologies that bridge the barriers to cross-
departmental knowledge access. More specifically, the findings suggest that cross-
departmental knowledge is more highly valued when technology features help
individuals find diverse opinions, see how knowledge about a situation has evolved
over time, link general overviews of knowledge about a situation to additional
related details, and actively add to the knowledge as it evolves.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER 1: EXECUTIVE SUMMARY
This is an age where organizations are fighting to maintain competitive
advantage. As such, it has become increasingly important to figure out ways to
influence employee innovation. Enabling employees to acquire knowledge from
other areas of expertise within the organization offers one alternative. Within the
organization, this often means acquiring knowledge from colleagues in other
departments. However, a major challenge organizations face is bridging the barriers
to cross-departmental knowledge access. It has been suggested that the barriers
lower individuals’ perceived value of cross-departmental knowledge on innovation.
Of particular interest to this dissertation study is bridging the barriers to cross-
departmental knowledge access through technology features. More specifically,
using technology features that support cross-departmental understanding is expected
to positively relate to individuals’ perceived value of cross-departmental knowledge
on innovation.
In this executive summary, the motivation for this dissertation study is
highlighted (section 1.1). This includes a discussion about the perceived value of
cross-departmental knowledge on innovation. Next, the conceptual model and
research hypotheses are outlined (section 1.2). This is followed by a summary of
this dissertation study’s research methodology (section 1.3), findings (section 1.4),
and contributions (section 1.5). The executive summary ends with an outline of the
remaining chapters (section 1.6).
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
2
1.1 Motivation
Innovation is comprised of divergent thinking and convergent thinking.
Divergent thinking is characterized as the synthesis of various disparate thoughts that
help generate ideas. In contrast, convergent thinking has been described as
analytical, evaluative thought that has a more solution-oriented focus. In the context
of this study, individuals’ perceived value of cross-departmental knowledge on
innovation refers to individuals’ perceived value of cross-departmental knowledge
on divergent thinking and convergent thinking. In other words, whether individuals
believe that cross-departmental knowledge helps their divergent thinking and
convergent thinking is considered. A better understanding of the factors that bridge
the barriers encountered during cross-departmental knowledge access, or otherwise
contribute to the perceived value of cross-departmental knowledge on innovation
might have profound implications for organizations wanting to influence employee
innovation. Yet, there has not been much research to determine these factors of
influence, particularly when it comes to the use of supportive technology features.
This is the goal of this dissertation study.
From prior research, some factors may appear to be obvious influencers, such
as perceived openness of communication, people-to-people interaction, perceived
task complexity, and perceived ease of use of technology. Less research has directly
examined the potential influence of supportive technology features, for example
those that support cross-departmental understanding. This leads to the major
research question under study. Does the use of technology features that support
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
3
cross-departmental understanding positively relate to individuals’ perceived value of
cross-departmental knowledge on innovation, after controlling for other factors in the
work environment including: perceived openness of communication, people-to-
people interaction, perceived task complexity, perceived ease of use of technology?
1.2 Overview of Conceptual Model and Hypotheses
In answering the research question posed in this dissertation study,
technology features that support cross-departmental understanding had to be
conceptualized. The theory of thought worlds was coupled with the theory of
hermeneutic inquiry to delineate a set of supportive technology features. According
to the theory of thought worlds, the barriers individuals encounter when accessing
cross-departmental knowledge are due to the fact that individuals are accessing
diverse funds of knowledge that have different systems of meaning. Not to mention
a readiness for directed perception that is entrenched in well-established skills and
experiences. However, from the theory it is also inferred that the barriers to cross-
departmental knowledge access can be bridged if the knowledge being accessed is:
a) associated with the expert source, b) interpretable, and c) adaptable.
The theory of hermeneutic inquiry describes a process by which individuals
come to understand, reflect on, and change their own interpretations of situations by
accessing the interpretations of others. The theory of hermeneutic inquiry goes
beyond the theory of thought worlds to describe several technology design principles
- ownership, multiplicity, easy travel, indeterminacy, and emergence - that support a
process of ongoing knowledge access among individuals. Collectively these
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
4
technology design principles are used to delineate the set of technology features
support cross-departmental understanding. The features bridge the barriers of cross-
departmental knowledge access because they enable access to knowledge that is: a)
associated with the expert source, b) interpretable, and c) adaptable. For example,
the principles of ownership and multiplicity maintain that individuals should author
their own knowledge based on their personal skills and experiences. This supports
access to knowledge that is associated with the expert source. By enabling
individuals to move between general knowledge and the associated contextual
details, the principle of easy travel supports access to knowledge that is interpretable.
By allowing individuals to flexibly change and track changes to knowledge on the
fly, the principles of indeterminacy and emergence provide access to knowledge that
is adaptable.
In short, individuals accessing cross-departmental knowledge encounter
barriers. Drawing from the theory of thought worlds, providing access to knowledge
that is: a) associated with the expert source, b) interpretable, and c) adaptable can
bridge the barriers. Drawing from the theory of hermeneutic inquiry, the use of
technology features that support cross-departmental understanding can provide
access to this knowledge. Using technology features in this manner is expected to
positively relate to the perceived value of cross-departmental knowledge on
innovation (i.e. divergent thinking and convergent thinking), after controlling for
other factors in the work environment, including: perceived openness of
communication, people-to-people interaction, perceived task complexity, perceived
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
5
ease of use of technology. The hypotheses that are posed are depicted in a
replication of Figure 3.2.
Figure 3.2 Conceptual Model
H i
H2
Control Variables:
+ Perceived openness of communication
+ People to people interaction
+ Perceived task complexity
Use of technology features
that support cross-
departmental understanding
Perceived value of cross-
departmental knowledge on
divergent thinking
Perceived value of cross-
departmental knowledge on
convergent thinking
+ Perceived ease of use of technology
1.3 Overview of Research Methodology
Data collection took place at Company A, a medium-sized (approximately
3,000 employees) scientific and technical organization, with a mission to prevent and
solve problems associated with the design, deployment, and operation of large,
highly complex space systems. This research site was chosen because employees at
Company A depended on cross-departmental knowledge when solving their
problems, their problems required innovation, and there were many KM technologies
and features available to the employees. This dissertation study proceeded in two
phases. The first phase, preliminary data collection, took place one year prior to
final survey administration to learn more about the current state of affairs with
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
6
respect to employees accessing each other’s knowledge. The second phase, a two-
wave survey, was used to test the hypotheses.
Preliminary data collection included participant observation of knowledge
management (KM) committee meetings, interviews with top management, and semi
structured case studies with the scientists and engineers who prevent and solve the
problems. The management perspective provided insight into the organizational and
technological support available to employees accessing each other’s knowledge.
Case studies verified that Company A employees did bridge cross-departmental
barriers by accessing knowledge that was: a) associated with the expert source, b)
interpretable, and c) adaptable. Preliminary data collection also confirmed that
technology features existed to help bridge cross-departmental barriers, and other
factors in the work environment (e.g. perceived openness of communication, people-
to-people interaction, perceived task complexity, perceived ease of use of
technology) should be considered.
Next, a two-wave web-based survey was administered. Two waves were
used to reduce common method variance bias and show discriminant validity among
the variables. The first wave survey took place in the Fall 2003. Data were collected
for the independent and dependent variables - perceived openness of
communication, people-to-people interaction, perceived task complexity, perceived
ease of use of technology, perceived value of cross-departmental knowledge on
innovation (i.e. divergent thinking and convergent thinking). Of the 874 employees
at Company A invited to participate in the first wave, 248 responses were available
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
7
for analysis (28.4%). The second wave survey occurred in the Spring 2004 to collect
dependent variable data only - perceived value of cross-departmental knowledge on
innovation (i.e. divergent thinking and convergent thinking). Of the 248 invited to
participate in the second wave, 131 responded (52.8%).
1.4 Summary of Findings
In the first wave data analysis, there was a lack of discriminant validity
among all of the constructs. This may have been due in part to common method
variance. The second wave data collection reduced common method variance bias
and showed that there was discriminant validity among all constructs. Regression
analysis indicated support for hypothesis 1 and 2: the use of technology features that
support cross-departmental understanding was positively associated with both the
perceived value of cross-departmental knowledge on divergent thinking and the
perceived value of cross-departmental knowledge on convergent thinking. In
addition, perceived ease of use of technology was positively related to the perceived
value of cross-departmental knowledge on convergent thinking. People-to-people
interaction was also shown to positively relate to the perceived value of cross-
departmental knowledge on divergent thinking and convergent thinking, albeit to a
lesser degree.
1.5 Summary of Contributions
Finding that the use of technology features that support cross-departmental
understanding was positively associated with the perceived value of cross-
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
8
departmental knowledge on divergent thinking and convergent thinking, provides
several contributions to theory, future research, and practice. One contribution to
theory is that this dissertation study showed that KM technologies could be designed
to facilitate knowledge access across organizational boundaries. More specifically,
findings suggest that cross-departmental knowledge is more valued when technology
features support knowledge that: allows individuals to make their biases known,
provides links to additional contextual details, provides contradictory alternatives
rather than a well formulated response, and is open to unpredictable change. This
dissertation study also showed that individuals value the use of KM technologies for
accessing cross-departmental knowledge, even when given the opportunity to use
people-to-people interaction.
In light of the findings from this dissertation study, future research might
consider how packaging the technology features might influence the perceived value
of cross-departmental knowledge on innovation. For example, Company A
employees utilized many different KM technologies to facilitate the use of
technology features that support cross-departmental understanding. Whether the use
of one KM technology with an integrated set of technology features is better for
innovation than a non-integrated best of breed KM technology solution is an
interesting question for future research. Company A employees also required access
to a lot of contextual details when accessing cross-departmental knowledge.
Determining the difference between too much and not enough contextual detail when
it comes to information overload and innovation is also an interesting question for
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
9
future study. Future research that considers the differential impact of self-managed
vs. technology-managed cross-departmental knowledge access on innovation would
also be a fruitful area of study. In addition, it would be interesting future research to
determine whether using technology features that support cross-departmental
understanding facilitates individuals’ intellectual development (e.g. know-what,
know-how, know-why) in these other areas of expertise.
When it comes to practice, the findings from this dissertation study suggest
how technology features might provide a means to address the challenge of getting
employees to seek, share, and otherwise transfer knowledge across organizational
boundaries. More specifically, organizations should evaluate technology features
based on whether they can help bridge cross-departmental barriers. For example,
technology features that allow for periodic updates may not be enough. Individuals
may be better off with technology features that support timely, personalized, alerts to
knowledge change that results from its ongoing evolution. In addition, organizations
should not look to implement technology features that sanitize knowledge to remove
biases and contrasting, inconsistent points of view. Individuals need to be able to
use technology features that can accommodate the contradictions that result from the
variety of personal views and opinions and help identify and resolve them. Lastly,
organizations should not look for technology features that reduce information
overload. Individuals need to use technology features that provide a means to
manage the relationships that result from accessing knowledge at various levels of
detail.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
10
1.6 Roadmap to the Dissertation
Chapter 2 introduces the dissertation study by providing a statement of the
problem and discussing background literature from innovation, creativity, KM, and
information system (IS) research. Chapter 3 uses the theory of thought worlds and
the theory of hermeneutic inquiry to describe the conceptual model and hypotheses.
The research methodology is described in Chapter 4, and includes a detailed
discussion of the preliminary data collection methods and survey methodology. The
results are presented in Chapter 5. Chapter 6 details the contributions to theory,
future research, and practice.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
11
CHAPTER 2: INTRODUCTION
Organizations are under more pressure today then ever before when it comes
to sustaining a competitive advantage in their respective industries. A major source
of competitive advantage for most organizations comes from whether they can
influence employee innovation (e.g. Argote & Ingram, 2000; Grant, 1996; Hargadon
& Sutton, 1997; Kogut & Zander, 1992; Leonard & Sensiper, 1998). Research
suggests that an organization that enables its employees to access each other’s non-
redundant (i.e. diverse) knowledge can positively influence their innovation (Hansen,
1999; Hargadon & Sutton, 1997; Kanter, 1988; Leonard & Sensiper, 1998;
Majchrzak, Cooper, & Neece, 2004; von Krogh, 1998). Within an organization, this
often means facilitating access to knowledge outside of the employees’ own
department or area of expertise. However, organizations face continued difficulties
when it comes to getting employees to seek, share, adopt, or otherwise transfer each
other’s knowledge across these types of organizational boundaries (Carlile, 2002;
Markus, 2001; Markus, Majchrzak, & Gasser, 2002; Quinn, Anderson, &
Finkelstein, 1996; Szulanski, 1996; von Hippie, 1994).
Findings from one stream of research suggest that individuals encounter
barriers when accessing knowledge outside of their own department, which lower
their perceived value of cross-departmental knowledge on innovation (Dougherty,
1992). While these findings have important implications for the innovation,
creativity, KM, and IS literatures, the phenomena have not received much research
attention. Specifically, further study of possible antecedent factors that bridge the
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
12
barriers to cross-departmental knowledge access, or otherwise positively relate to the
perceived value of cross-departmental knowledge on innovation, remains nascent.
Take technology as an example. Organizations have been offering various
technological solutions to help employees access each other’s knowledge for
decades. Moreover, one of the strengths of technology is its ability to widely
disseminate diverse knowledge for access (e.g. Alavi & Leidner, 2001; Huber,
1990). Yet, little is known about the influence supportive technologies might have
on individuals’ perceived value of cross-departmental knowledge on innovation.
This area of research is of particular interest in this dissertation study.
This chapter begins by describing innovation (section 2.1) and noting that the
perceived value of cross-departmental knowledge on innovation is a critical area of
research that has not received much attention (section 2.2). This is followed by a
discussion of several factors expected to positively influence individuals’ perceived
value of cross-departmental knowledge on innovation (section 2.3). A review of the
current IS literature highlights the need for further study of the relationship between
the use of supportive technology features and the perceived value of cross-
departmental knowledge on innovation (section 2.4). The chapter ends with a
summary, including a recap of the major research question under study (section 2.5).
2.1 Innovation
According to Amabile et al. (1996), innovation starts with the production of
novel useful ideas (i.e. creativity) and ends with their successful implementation.
Scholars contend that it is comprised of divergent thinking and convergent thinking
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
(e.g. Basadur, 1994; Brophy, 1998; Couger, 1996; Leonard & Sensiper, 1998 p. 116;
Runco, 1994; Treffinger, Isaksen, & Dorval, 1994). Divergent thinking has been
described as thinking in various directions (Runco, 1999). It is characterized by
ideation and the originality and diversity of thought (Basadur, 1994; Brophy, 1998;
Couger, 1996). It is defined as the combination of multiple, unrelated matrices of
thought that generate novel ideas (Jabri, 1991; Leonard & Sensiper, 1998). In
contrast, convergent thinking has been described as having a more focused, linear
flow (Jabri, 1991). It is characterized by the evaluation and implementation of
potential solutions (Basadur, 1994; Couger, 1996; Treffinger et al., 1994). It is
defined as methodical, analytical, evaluative thought that permits a more narrowed
solution focus (Jabri, 1991; Leonard & Sensiper, 1998).
Some research has equated divergent thinking and convergent thinking with
individuals’ innate skills and abilities (e.g. Garfield, Taylor, Dennis, & Satzinger,
2001; Runco, 1999). In other words individuals are bom as divergent thinkers or
convergent thinkers. Other research has equated divergent thinking and convergent
thinking to individuals’ preferences and suggests that the preferences can be
influenced (e.g. Amabile, 1988; Couger, 1996; Jabri, 1991). In other words,
everyone has the capability to think in a divergent and/or convergent manner given
the proper conditions (Amabile, 1988; Couger, 1996; Treffinger et al., 1994). While
the research from both perspectives is insightful, this dissertation study takes the
latter view. Thus, in the context of this dissertation study, the perceived value of
cross-departmental knowledge on innovation refers to the perceived value of cross-
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
14
departmental knowledge on divergent thinking and convergent thinking. In other
words, this dissertation study considered how individuals’ perceived cross-
departmental knowledge helped their divergent thinking and convergent thinking.
2.2 Perceived Value of Cross-departmental Knowledge on Innovation
Creativity and innovation literature acknowledges that individuals’
innovation can flourish given access to a requisite variety of knowledge from other
people (Amabile, 1988; Garfield et al., 2001; Grant, 1996; Kanter, 1988; Leonard &
Sensiper, 1998). Yet, research has found that individuals charged to be innovative
(e.g. new product developers), do not always take advantage of the diversity of
knowledge from other employees that has been made available for access
(Dougherty, 1992; Markus et al., 2002). Prior research has addressed this issue by
examining the factors associated with the costs and difficulties of transferring
knowledge from other people across departments, areas of expertise, or other
organizational boundaries (Markus, 2001; Szulanski, 1996; von Hippie, 1994). Less
research has turned the problem around to consider factors that influence individuals'
perceived value of cross-departmental knowledge on innovation. Even less has
considered the contributions the use of supportive technologies might make.
This dissertation study addresses the last two points. According to the theory
of thought worlds, individuals encounter barriers when accessing knowledge outside
of their own department or area of expertise. The barriers lower the individuals’
perceived value of cross-departmental knowledge on innovation (Dougherty, 1992).
For example, individuals were found to inadvertently filter out knowledge from
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
15
people in other departments within the organization when they could not readily
understand the different interpretations at play (Dougherty, 1992). Not being able to
recognize, understand, and evaluate the different points of view presented a barrier
that diminished the individuals’ perceived value of cross-departmental knowledge on
innovation.
However, the theory of thought worlds and related research (e.g. Carlile,
2002; Majchrzak et al., 2004; Markus, 2001) also suggests that the barriers
individuals encounter when accessing knowledge outside of their own department
can be bridged, if the individuals are provided access to knowledge that is: a)
associated with the expert source, b) interpretable, and c) adaptable. In turn,
individuals’ perceived value of cross-departmental knowledge on innovation is
expected to be positively influenced.
2.3 Influencing Factors
The aim of this dissertation study is to consider the factors that support access
to knowledge that is: a) associated with the expert source, b) interpretable, and c)
adaptable, or otherwise influence the perceived value of cross-departmental
knowledge on innovation. When it comes to the various KM technologies available
in the marketplace, some scholars have argued that the technologies cannot possibly
support the processes of interpretation and adaptation individuals require when
accessing diverse knowledge from people in other departments or functional areas
(Dixon, 2000; Hansen, Nohria, & Tierney, 1999; Nonaka & Takeuchi, 1995; ODell
& Grayson, 1998). Other scholars have argued that the perceived inadequacies of
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
16
technological support stem from a lack of research when it comes to how the design
and use of technology features might differ based on the type of knowledge user or
the purpose for which the knowledge is used (Alavi & Leidner, 2001; Boland,
Tenkasi, & Te'eni, 1994; Markus, 2001; Mason & Mitroff, 1973). Drawing from the
theory of hermeneutic theory, the use of technology features that support cross-
departmental understanding is of particular interest in this dissertation study. More
specifically, whether a positive relationship exists between individuals’ use of
technology features that support cross-departmental understanding and their
perceived value of cross-departmental knowledge on innovation (i.e. divergent
thinking and convergent thinking) is examined.
Other factors believed to bridge the cross-departmental barriers are perceived
openness of communication and people-to-people interaction. For example, research
suggests that individuals working in a communicative environment may be more
willing to help each other interpret, and adapt each other’s knowledge (Andrews &
Delahaye, 2000; Borgatti & Cross, 2003; Szulanski, 1996; von Krogh, 1998). In
addition, increasing the opportunity for people-to-people interaction (e.g. chance
meetings, corporate training) has been suggested as a means to support the rich
communication needed to bridge the barriers associated with accessing other
people’s diverse knowledge (Hansen et al., 1999; Hargadon & Sutton, 1997; Nonaka
& Takeuchi, 1995; O'Dell & Grayson, 1998). In short, given perceived openness of
communication and people-to-people interaction, the contribution of knowledge
from other departments is more likely to be identified, understood, and used.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
17
Still other research suggests that perceived task complexity and perceived
ease of use of technology might positively relate to individuals’ perceived value of
cross-departmental knowledge on innovation. Bounded rationally, individuals lack
the diverse skills and experiences required to solve complex tasks, and thus, must
rely on the knowledge of others (Grant, 1996; Gray, 2000; Kanter, 1988; Leonard &
Sensiper, 1998; Simon, 1991). Under such circumstances, they are also more likely
to perceive the value cross-departmental knowledge brings to their work. Perceived
ease of use of technology may be another influencing factor. Research found that
individuals who accessed information through technology and believed the
technology was easy to use, also believed using the technology enhanced their
learning (Vandenbosch & Higgins, 1995). This leads to the first research question.
Research Question: Does the use of technology features that
support cross-departmental understanding positively relate to
individuals’ perceived value of cross-departmental knowledge
on innovation, after controlling for other factors in the work
environment including: perceived openness of communication,
people-to-people interaction, perceived task complexity, and
perceived ease of use of technology?
2.4 Technology Features That Bridge Thought World Barriers
In order to address the research question, a viable use of supportive
technology features has to be conceptualized. Drawing from Dougherty’s (1992) use
of the theory of thought worlds, knowledge that is: a) associated with the expert
source, b) interpretable, and c) adaptable will bridge the barriers individuals
encounter when accessing knowledge outside their own department. While
Dougherty (1992) suggests ways organizations can support access to knowledge that
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
18
bridges cross-departmental barriers, her primary focus is on cultural solutions, not
technological ones. Turning to the IS literature, research on supportive technology
features that provide access to knowledge that bridges cross-departmental barriers is
also limited.
Prior IS research has tended to focus on technologies that support a well-
planned set of steps or a uniform user base (Alavi & Leidner, 2001; Boland et al.,
1994; Markus et al., 2002; Mason & Mitroff, 1973). These traditional
conceptualizations of technology design and use can be seen in research on decision
support systems and expert systems. These rule-based systems are designed to
support structured problems that recur within a particular functional area (e.g.
Sprague Jr., 1980). As such, these systems are not designed to address cross-
disciplinary problems or cater to a diverse user base (El Sherif & El Sawy, 1988).
While the expert systems are used to access expertise, they are built to support the
less proficient workers (i.e. novices) within the particular knowledge domain, not
those outside of it (Yoon, Guimaraes, & O'Neal, 1995). Group support systems have
the capability to provide access to knowledge derived from a diverse set of known
experts. However, the primary emphasis of group support systems research has been
on helping people manage meetings and build consensus (e.g. Benbasat & Lim,
1993; Nunamaker Jr., Briggs, Mittleman, Vogel, & Balthazard, 1997; Zigurs &
Buckland, 1998), not interpret and adapt the knowledge they are exchanging (Boland
et al., 1994; Satzinger, Garfield, & Nagasundaram, 1999). Prior research on
corporate repositories and databases also has shortcomings, because the success of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
these technologies has been limited to pockets of users that work in the same rather
than different functional areas (Markus, 2001). Corporate electronic directories have
been suggested as a means to identify experts (e.g. Ruggles, 1998). However, their
technology features do not provide direct access to the experts’ knowledge.
There are scholars who have argued that technology can support more than a
well-planned set of steps and a uniform user base (e.g. Boland et al., 1994; Mason &
Mitroff, 1973). Yet, reviewing prior IS literature, it is evident that IS researchers
have not given much consideration to supportive technology features that provide
access to knowledge that is: 1) associated with the expert source, 2) interpretable,
and 3) adaptable. More recently, scholars have argued that the IS community must
reconceptualize the traditional ideas it has about the design and use of the many KM
technologies at our disposal (Alavi & Leidner, 2001; Markus, 2001). In a recent
review of KM related research, Markus (2001) noted some critical differences
between traditional forms of KM technology support and KM technology that
supports cross-departmental knowledge access, and urged researchers to start
acknowledging and addressing the differences in their studies.
Boland et al. (1994) provide one of the few research discussions that offer
considerable detail as to how technology features can be designed to support an
individual’s need to understand knowledge from other people. Using the theory of
hermeneutic inquiry, the authors describe a process by which individuals understand,
reflect on, and change their own interpretations of situations by accessing the
interpretations of others through technology. They also outline technology design
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 0
principles that support the process. In this dissertation study, the theory of
hermeneutic inquiry and the accompanying design principles are coupled with the
theory of thought worlds to outline a set of technology features to support the access
to knowledge that is: 1) associated with the expert source, 2) interpretable, 3)
adaptable. Collectively the set of technology features is conceptualized as the use of
technology features that support cross-departmental understanding. Using
technology in this manner is expected to positively relate to individuals’ perceived
value of cross-departmental knowledge on innovation.
2.5 Summary
In sum, an organization can sustain competitive advantage by influencing
employee innovation. Enabling organizational members to effectively access
knowledge from other areas of expertise is one alternative. Within an organization,
this means accessing knowledge from other departments, which presents barriers that
lower individuals’ perceived value of cross-departmental knowledge on innovation.
However, providing individuals access to knowledge that is: a) associated with the
expert source, b) interpretable, and c) adaptable, can bridge these barriers. As a
result, individuals’ perceived value of cross-departmental knowledge on innovation
is expected to increase. From prior research, it is suggested that perceived openness
of communication, people-to-people interaction, perceived task complexity, and
perceived ease of use of technology will positively relate to individuals’ perceived
value of cross-departmental knowledge on innovation. Whether the use of
technology features that support cross-departmental understanding follows suit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
remains unsettled. To provide further insight, this dissertation study poses the
following research question.
Does the use of technology features that support cross-
departmental understanding positively relate to individuals’
perceived value of cross-departmental knowledge on
innovation, after controlling for other factors in the work
environment, including: perceived openness of communication,
people-to-people interaction, perceived task complexity, and
perceived ease of use of technology?
The next chapter explains the theory of thought worlds and the theory of
hermeneutic inquiry in more detail. A set of technology features that support cross-
departmental understanding is also described. This is followed by a presentation of
the conceptual model and hypotheses. The chapter ends after the presentation of an
exploratory research question.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2
CHAPTER 3: CONCEPTUAL MODEL
This chapter opens with a presentation of the theory of thought worlds. An
explanation of the barriers individuals encounter when accessing knowledge from
different departmental thought worlds is outlined (section 3.1). This includes a
description of how the barriers impair individuals’ perceived value of cross-
departmental knowledge on innovation. Further elaboration of the theory suggests
that individuals who can access knowledge that is: a) associated with the expert
source, b) interpretable, and c) adaptable, can bridge the barriers, and thereby
increase their perceived value of cross-departmental knowledge on innovation
(section 3.2). While applications of the theory have provided different solutions that
provide access to knowledge that supports cross-departmental bridging,
technological solutions have not been offered (section 3.3). Drawing from the theory
of hermeneutic inquiry, and the technology design principles that support it, a set of
technology features that support cross-departmental understanding is outlined
(sections 3.4 and 3.5). Using both theories, it is hypothesized that the use of
technology features that support cross-departmental understanding is positively
related to individuals’ perceived value of cross-departmental knowledge on
innovation (section 3.6). The chapter ends with an exploratory research question
(section 3.7). The question considers whether there are interaction effects between
the use of technology features that support cross-departmental understanding and
other factors in the work environment, such as perceived openness of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3
communication, people-to-people interaction, perceived task complexity, and
perceived ease of use of technology.
3.1 Theory of Thought Worlds
Innovation requires access to insights from a variety of knowledge
specialties. Dougherty (1992) calls these knowledge specialties thought worlds. “A
thought world is a community of persons engaged in a certain domain activity who
have a shared understanding about that activity” (Dougherty, 1992 p. 182).
Dougherty (1992) focuses on thought worlds within an organization that are divided
along departmental boundaries, such as manufacturing, marketing, sales, and
technical departments. Although similar to Lave and Wenger’s (1991) notion of
communities of practice, the theory of thought worlds focuses on accessing
knowledge across, rather than within community boundaries. Dougherty’s (1992)
work is provocative, because it shows that organizations trying to influence their
employees’ innovation cannot assume their employees value knowledge outside of
their immediate department. Given ready access to employee knowledge from other
departments, during a project where innovation was the focal point, Dougherty
(1992) found individuals only focused on how their own area of expertise could be
used to complete a task. According to Dougherty (1992), this is due in part to the
barriers individuals encounter when accessing knowledge from different thought
worlds. The barriers are said to lower individuals’ perceived value of cross-
departmental knowledge on innovation. According to the theory, the barriers result
from the different a) funds of knowledge, b) systems of meaning, and c) readiness
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4
for directed perception that comprise each thought world. In the following
paragraphs, the theory of thought worlds is drawn upon to discuss these three
concepts. The potential benefit each of these elements brings to innovation when
accessing knowledge from different thought worlds (i.e. departments) is also
discussed. The barriers individuals encounter, which lower their perceived value of
cross-departmental knowledge on innovation, are described as well. Supporting
evidence found in the innovation and KM literatures adds to the discussion.
3.1.1 Funds of Knowledge
According to Dougherty (1992), each thought world represents a different
fund of knowledge (i.e. people know different things). A fund of knowledge
represents an interest or area of expertise that is rooted in the skills and experiences
members of a thought world have developed over time (Dougherty, 1992). A diverse
set of funds is valued for innovation, because non-redundant knowledge is more
likely to become available. As such, there is a greater potential to access alternative
ideas and encounter new and different opportunities for knowledge use that
contribute to innovation (Hansen, 1999; Hargadon & Sutton, 1997; Kanter, 1988;
Leonard & Sensiper, 1998; Majchrzak et al., 2004). However, accessing diverse
knowledge outside of one’s area of expertise can also be problematic, because
individuals lack suitable criteria to judge whether the knowledge is appropriate in the
context of their current work (Markus, 2001). Without the proper cues that indicate
the credibility, usefulness, and quality of knowledge, (Borgatti & Cross, 2003;
Majchrzak et al., 2004; Sussman & Siegal, 2003), it may go unused.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5
3.1.2 Systems of Meaning
According to Dougherty (1992), each of the thought worlds also has a
different system of meaning (i.e. people know things differently). This presents
another barrier, in that people belonging to different thought worlds often interpret
the same situation differently (Dougherty, 1992). Providing access to different
thought worlds that produce qualitatively different understandings is good for
innovation, because the contrasts challenge individuals’ personal beliefs and
assumptions, and force them to think outside the box (Boland & Tenkasi, 1995;
Kanter, 1988). However, research shows that access to knowledge from different
thought worlds can also be a barrier, in absence of support to recognize, understand,
and carefully consider the different points of view at play (Boland & Tenkasi, 1995;
Hansen, 1999; Markus et al., 2002; Szulanski, 1996; von Hippie, 1994). Dougherty
(1992) found that individuals had trouble identifying the potential contributions
knowledge from different thought worlds could make; as a result knowledge that was
critical to their innovation was inadvertently filtered out.
3.1.3 Readiness for Directed Perception
Each thought world also develops a readiness for directed perception.
Dougherty (1992) describes it as an intrinsic harmony, where members of a thought
world develop a particular point of view based on the expertise they and other
members of their thought world have developed over time. Access to well-
established skills and experiences makes for more insightful knowledge
contributions that lead to innovation (Nonaka & Takeuchi, 1995). However,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6
knowledge from different thought worlds operates under different systems of
meaning, and inconsistencies arise, which upsets individuals’ intrinsic harmony
(Dougherty, 1992). If individuals accessing knowledge from different thought
worlds accept the inconsistencies, they alter their own readiness of directed
perception that they have come to value (Carlile, 2002). This is seen as a negative
consequence, and leads individuals to deliberately reject the knowledge, even though
it may have contributed to their innovation (e.g. Dougherty, 1992).
3.2. Accessing Knowledge That Bridges Thought World Barriers
Drawing from the theory of thought worlds, different funds of knowledge,
systems of meaning, and readiness for directed perception create barriers to
accessing cross-departmental knowledge. In turn, individuals’ perceived value of
cross-departmental knowledge on innovation is lowered. However, further
elaboration of the theory suggests that accessing knowledge that is: a) associated
with the expert source, b) interpretable, and c) adaptable can bridge the barriers. In
turn, individuals’ perceived value of cross-departmental knowledge on innovation is
expected to increase. The following paragraphs describe how accessing knowledge
in this manner bridges the barriers to cross-departmental knowledge access and
positively influences individuals’ perceived value of cross-departmental knowledge
on innovation. Prior research in the innovation, KM, organizational learning,
boundary object, and IS literatures is drawn from to support these assertions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7
3.2.1 Knowledge Should Be Associated With the Expert Source
The theory of thought worlds suggests that access to a variety of funds of
knowledge contributes to innovation. However, individuals accessing knowledge
outside of their area of expertise often lack criteria to judge the quality of the
knowledge. Thus, knowledge should be associated with the expert source. By
identifying expert sources of the knowledge, research has found that the reputation of
the experts can contribute to favorable assessments of the knowledge (Andrews &
Delahaye, 2000; Majchrzak et al., 2004; Orlikowski, 2000). The credibility of the
source is used as a cue for the credibility of the knowledge (e.g. Sussman & Siegal,
2003). Individuals also believe that they have better access to expert thinking when
they can identify the experts of the knowledge (Borgatti & Cross, 2003). They are
likely to believe the knowledge is more useful, because they have a means to access
additional details about the knowledge that might help them solve their problem (e.g.
Borgatti & Cross, 2003; Majchrzak et al., 2004).
3.2.2 Knowledge Should be Interpretable
Second knowledge should be interpretable. This follows from the barrier
associated with the different systems of meaning individuals encounter when
accessing knowledge outside their department. Knowledge accessed outside the
department should be interpretable, such that people outside the department can
understand the meaning people inside the department give it. Conventional wisdom
has led us to believe that individuals accessing knowledge outside of their own
department will readily understand and appreciate abstract knowledge. However,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8
individuals with access to both abstract, general knowledge, and specific, concrete,
contextual details engage in a more comprehensive understanding and analysis of
others’ knowledge (Brown & Duguid, 1991; Majchrzak et al., 2004; Nonaka &
Takeuchi, 1995). Individuals are also more likely to benefit from the contrasting
points of view held by others, when given access to the contextual details that
support the rationale behind the views (Boland & Tenkasi, 1995; Markus, 2001;
Markus et al., 2002).
3.2.3 Knowledge Should Be Adaptable
Third knowledge should be adaptable. This addresses the barrier associated
with the readiness of directed perception. With malleable knowledge, individuals are
more likely to compare, contrast, and integrate their own point of view with the
points of view of other people from other areas of expertise (Henderson, 1991; Star
& Griesemer, 1989). Thus appreciation, rather than outright rejection of knowledge
from other departments is more likely (Carlile, 2002; Dougherty, 1992). In addition,
the inconsistencies individuals encounter when accessing knowledge outside of their
own department are more likely to be considered, because the individuals are free to
reconcile those that upset their intrinsic harmony (Carlile, 2002). Innovation rests
not on how well knowledge of the past can be reapplied as is, but on how well
knowledge of the past can be adapted to fit current conditions (Hargadon & Sutton,
1997; Walsh & Ungson, 1991). As such, people are more likely to value knowledge
when they are free to make new interpretations of it, in light of their current skills
and experiences. Moreover, knowledge that is undergoing dynamic change over
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9
time and across contexts is more likely to be valued, when the changes are being
preserved and developed as it evolves (e.g. El Sawy & Bowles, 1997; Stein &
Zwass, 1995).
3.3 Cultural Support That Bridges Thought World Barriers
Dougherty (1992) provides several ideas about how to support access to
knowledge that helps bridge barriers to cross-departmental knowledge access.
However, the descriptions of support are broad and only focus on cultural solutions.
For example, it is suggested that individuals will have greater appreciation for the
unique insights from different funds of knowledge, if they develop a plan to
deliberately and directly address the differences as they come up. It is also suggested
that interactions based on appreciation and joint development allow individuals to
value the contrasting perspectives that would otherwise upset their intrinsic harmony.
Establishing interdisciplinary responsibility for tasks is suggested as a means to help
individuals understand and appreciate knowledge that is based on different systems
of meaning. While these ideas provide insight into how to bridge barriers of cross-
departmental knowledge access, they do not adequately address the needs of this
dissertation study. Given that the major aim of this study is to show that the barriers
to accessing cross-departmental knowledge can be bridged through the use of
supportive technology features, addressing cultural solutions is not of primary
interest.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0
3.4 Theory of Hermeneutic Inquiry
To consider technology features that support access to knowledge that is
associated with the expert source, interpretable, and adaptable, this dissertation study
draws from the theory of hermeneutic inquiry. “A hermeneutic process of inquiry
involves actors who make interpretations of situations and reflect upon their action
and their interpretations in order to push to the horizons of their understanding”
(Boland et al., 1994 p. 464). The theory is useful for this dissertation study, because
as the authors indicate, it describes how individuals, who act as autonomous agents,
come to understand, reflect on, and change their own interpretations by opening
themselves up to others who have different interpretations.
According to Boland et al. (1994), individuals use their interpretations as a
basis for understanding the world. Individuals’ interpretations are grounded in
traditions that influence how they see the world. The use of the term “traditions” is
similar to the term systems of meaning that is described in the theory of thought
worlds. Recognizing that individuals’ understanding is bound by tradition, the
theory of hermeneutic inquiry advocates that individuals access knowledge from
other people in order to push to the horizon of their understanding. According to
Boland et al. (1994), accessing other people’s knowledge enables individuals to be
open to the traditions of others, and reflect on their own and others understanding in
the context of the different traditions. The use of the phrase “push to the horizon of
understanding” is similar to the notion of altering one’s readiness of perception. The
theory of hermeneutic inquiry expands beyond the theory of thought worlds, because
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
it elaborates on how features of technology might support a process of ongoing
knowledge access among individuals. As described, the theory can be applied to
delineate technology features that support access to knowledge that helps bridge the
barriers associated with cross-departmental knowledge access.
More specifically, Boland et al. (1994) argue that individuals can make their
traditions and horizons of understanding visible through the use of supportive
technology features. According to Boland et al. (1994), internally consistent fact
nets are used to represent individuals’ knowledge. The authors go on to describe fact
nets as networks of contingent truths individuals develop to support their own
beliefs. Individuals use them to describe the relationships they believe exist among
entities, including causal relations, assumptions, constraints, and presuppositions
(Boland et al., 1994). According to the theory, all individuals must be able to make
their own independent representations of situations (e.g. fact nets) based on their
own skills and experiences. It is only when individuals are open to the various,
different horizons of understanding of other people that they are able to reflect on
and push their own horizon of understanding.
When providing access to a diverse set of other people’s representations,
thought also has to be given as to how individuals can understand these
representations. According to Boland et al. (1994), individuals are able to do so by
moving back and forth between different levels of the representations. Individuals
viewing someone else’s representations are more likely to understand the
representations when they are equally enabled to access a general overview of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2
representations, as well as the minute details (Boland et al., 1994). The theory goes
on to explain the goal of accessing other people’s knowledge. It is not to develop
one “true picture of a situation”, but rather come to an understanding that is useful to
the situation at hand (Boland et al., 1994). As such, representations should never be
considered complete or formalized to a point that they are unchangeable. Individuals
should have a means to configure and reconfigure understandings of a situation
(Boland et al., 1994). According to the authors, this includes keeping track of how
representations evolve over time and across contexts.
3.5 Technology Features that Support Cross-departmental Understanding
Drawing from the theory of thought worlds, providing access to knowledge
that is: 1) associated with the expert source, 2) interpretable, and 3) adaptable, helps
bridge the barriers individuals encounter when accessing cross-departmental
knowledge and thus positively relates to individuals’ perceived value of cross-
departmental knowledge on innovation. Drawing from the theory of hermeneutic
inquiry, Boland et al. (1994) offer several technology design principles - ownership,
multiplicity, easy travel, indeterminacy, emergence - to provide access to knowledge
that helps bridge the barriers to cross-departmental knowledge access. Table 3.1 lists
the design principles and their definitions. In the following paragraphs each design
principle is discussed, including how it bridges the cross-departmental barriers. How
it contributes to the perceived value of cross-departmental knowledge on innovation
is also delineated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3
Table 3.1. Technology Design Principles
Technology Design
Principles
Descriptions
Ownership Identifies experts of knowledge based on existing skills and experiences.
Multiplicity Provides multiple interpretations of the same or similar situations from different points of
view.
Easy Travel Enables movement across different levels of detail within and across knowledge
representations.
Indeterminacy Leaves knowledge representations open to interpretation.
Emergence Creates and refines knowledge categories and constructs and levels of abstraction.
3.5.1 Supporting Associations Between Knowledge and the Expert Source
First, individuals should have access to knowledge that is associated with the
expert source. According to the theory of hermeneutic inquiry, representations that
provide access to traditions and horizons of understanding can only be provided at
the level of the individual (Boland et al., 1994). Thus, the principle of ownership
advocates that knowledge sources author their own independent representations of
situations based on their personal skills and experiences. Authorship links the
knowledge with a source, which provides cues about the source’s credibility that
help individuals make more informed decisions about the quality, relevance, and
usefulness of knowledge that is outside of their area of expertise (Markus, 2001;
Orlikowski, 2000; Sussman & Siegal, 2003). Identifying the source of the
knowledge also provides individuals with a means to gain access to the source’s
thinking, which helps individuals during solution implementation (Borgatti & Cross,
2003; Majchrzak et al., 2004). Being able to narrow down the selection of
knowledge to that which is most relevant and useful, and most likely to help solution
implementation is likely to contribute to individuals’ perceived value of cross-
departmental knowledge on convergent thinking.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 4
Given that expert sources have the opportunity to author their own
independent representations of situations based on their own personal skills and
experiences, the theory of hermeneutic inquiry also suggests that multiple
interpretations of situations from different points of view will be available for access
(Boland et al., 1994). This supports the principle of multiplicity, which is a likely
contributor to the perceived value of cross-departmental knowledge on divergent
thinking. Research has suggested that the variety of interpretations can be a source
of creative abrasion, such that the contrasting perspectives generate alternative ideas
(Boland & Tenkasi, 1995; Kanter, 1988; Leonard & Sensiper, 1998).
3.5.2 Supporting Interpretable Knowledge
Second, individuals should have access to knowledge that is interpretable.
The theory of hermeneutic inquiry suggests that the underlying structure of
understanding is comprised of a tacking back and forth between theory and details
within and across representations (Boland et al., 1994). This implies that any entity
can be linked to any other entity, such that individuals accessing knowledge from
other departments can have easy travel between different levels of a representation
(Boland et al., 1994). Easy travel between multiple levels of a representation allows
individuals to deliberate and reflect on each other’s knowledge (Mason & Mitroff,
1973). This is especially helpful when individuals are accessing knowledge from
other people who have beliefs and judgments that are different from their own
(Boland & Tenkasi, 1995; Boland et al., 1994; Markus, 2001; Mason & Mitroff,
1973). Deliberation and reflection are likely contributors to the perceived value of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 5
cross-departmental knowledge on convergent thinking. In addition, the ability to
travel across a varied set of representations enables individuals to recognize,
understand, and carefully consider the ways to trade-off and integrate knowledge
from the different interpretations at play (Boland & Tenkasi, 1995; Boland et al.,
1994; Brown & Duguid, 1991; Mason & Mitroff, 1973). These actions are likely to
contribute to the perceived value of cross-departmental knowledge on divergent
thinking.
3.5.3 Supporting Adaptable Knowledge
Third, individuals should have access to knowledge that is adaptable.
Similarly, the idea behind hermeneutic inquiry is to leave room for knowledge to
change and grow (Boland et al., 1994). Rather than achieve one acceptable,
enduring understanding of a situation, indeterminacy allows other people’s
representations to be considered incomplete and fuzzy (Boland et al., 1994).
Individuals are more likely to appreciate the contributions of other’s knowledge
when they believe that they can freely change and integrate it to suit their current
needs (Hargadon & Sutton, 1997; Majchrzak et al., 2004). Greater leeway in reusing
other people’s knowledge is likely to contribute to the perceived value of cross-
departmental knowledge on divergent thinking.
By supporting an ongoing access to other’s diverse knowledge, and a
continuous process of interpretation, it is expected that the reification of knowledge,
known to stifle innovation (Hargadon & Sutton, 1997; Walsh & Ungson, 1991), will
be avoided. As knowledge is accessed over time and across contexts, the emergence
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 6
of interpretations and representations will unfold (Boland et al., 1994). Detailed
relationships among entities will be refined, and new entities and relationships
among them will be created (Boland et al., 1994). As such, individuals are more
likely to appreciate knowledge when the representations are captured and tracked as
they evolve over time and across contexts. The constant recreation and refinement
of knowledge characteristic of indeterminacy and emergence is likely to contribute to
the perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking.
3.5.4 An Example of Technology Features that Support Cross-departmental
Understanding
To consider what individuals’ might experience through the use of
technology features that support cross-departmental understanding, consider the
following scenario. A launch vehicle fails to take a satellite into geosynchronous
orbit. Raymond, an expert in propulsion, is called to solve the problem. Using a
web-based interface, in one window, he starts by looking at the history of the launch
vehicle, including: the design review, technical specification, vehicle configuration,
and anomaly report. In another he searches for past satellite launch problems, where
the launch vehicle failed. He is provided a set of past problem solutions that fit the
criteria. The solutions are authored by experts in propulsion, orbit and mechanics,
and altitude control. With the set of past problem solutions, Raymond has several
linking options. He can: 1) compare the solutions based on their technical
narratives, which include, a statement of the problem, recommendations for solution,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 7
reproduction of failure, implementation procedures used to solve the problem, and
areas of expertise required to solve the problem, 2) link to more detailed views of an
individual problem solution, 3) link to prior solutions the authors used to solve their
problems, 3) link to other solutions authored by the same experts, 4) search for other
experts with the same areas of expertise, or 5) enter knowledge forums related to
propulsion, orbit and mechanics, and altitude control to discuss his problem with
others.
Raymond decides looks at the reproduction of failures across the set of
problem solutions. The one authored by the altitude control expert looks similar to
the situation he is facing. Next, he clicks on the altitude control expert’s solution
recommendation. One of the solution recommendations dealt with the calculations
that were being used. Raymond can read the authors’ explanation about the
calculation as well as look at the calculation in more detail. Clicking on one of the
calculations launches the modeling and simulation application that was used to run
the calculations. It is ready to use. The modeling and simulation tool is comprised
of the launch vehicle components, component parameters, and the relationships
among the components. Rolling the mouse over the components provides additional
contextual details that provide the rationale behind the expert’s decision to change
the calculations. The source code used to run the application is available for
download. If Raymond so chooses, he can change and recompile the source to meet
the needs of his current problem-solving context.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 8
3.5.5 Summary
In summary, the theory of hermeneutic inquiry describes how individuals can
use technology features to understand, reflect on, and change their own
interpretations, by accessing knowledge from other people who have different
interpretations. Understanding that individuals have different traditions and horizons
of understanding, the process of hermeneutic inquiry does not support the removal of
these biases, but instead supports a higher visibility of them. In sustaining this
process through the use of technology features that support cross-departmental
understanding the barriers to cross-departmental knowledge access are bridged.
3.6 Control Variables
From prior research, several factors are included in this dissertation study as
control variables, including a) perceived openness of communication, b) people-to-
people interaction c) perceived task complexity, and d) perceived ease of use of
technology. These factors have not been explicitly examined for their influence on
the perceived value of cross-departmental knowledge on innovation. However,
perceived openness of communication, people-to-people interaction, and perceived
task complexity are consistently found to positively contribute to creative and
innovative performances (Amabile et al., 1996; Dougherty, 1992; Hansen et al.,
1999; Hargadon & Sutton, 1997; Majchrzak et al., 2004; Nonaka & Takeuchi, 1995;
Oldham & Cummings, 1996). In addition, prior IS research suggests that the
perceived ease of use of technology may be as important as the use of technology
features that support cross-departmental understanding, when it comes to examining
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 9
the relationship between the use of technologies and the different forms of
technology acceptance (Davis, 1989). In the following paragraphs, each of these
four control variables is discussed in turn.
3.6.1 Perceived Openness of Communication
Perceived openness of communication describes the degree to which
individuals believe their communication with others will be easy (O'Reilly &
Roberts, 1977). Research suggests that individuals’ decision to access other’s
knowledge is influenced by the extent they believe that they are working in a
communicative environment (Goodman & Darr, 1998; Szulanski, 1996). Individuals
that feel comfortable offering and receiving each other’s help are more likely to
value the potential impact each other’s knowledge has on their performance. More
willing to communicate, individuals may also care more about understanding each
other’s perspectives (Amabile, 1988; von Krogh, 1998), be more open to evaluating
and resolving the inconsistencies inherent in diverse knowledge (Carlile, 2002;
Dougherty, 1992), and be more willing to figure out ways to adapt and integrate the
knowledge with their own (Majchrzak et al., 2004). As such, perceived openness of
communication is expected to positively relate to the perceived value of cross-
departmental knowledge on divergent thinking and convergent thinking.
3.6.2 Perceived Task Complexity
Perceived task complexity is characterized by the degree to which individuals
believe their work assignments do not have a clear, direct, or easily recognized
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 0
solution (George & Zhou, 2001). Research suggests that individuals who are
uncertain about what should be done to complete a task, or how they should go about
completing it, are confronted with an intellectual challenge (Amabile et al., 1996;
Oldham & Cummings, 1996). Bounded rationally and not able to solve the problem
on their own, they have to access knowledge from others outside of their area of
expertise (Gray, 2000; Simon, 1991). On the one hand, individuals working on
complex tasks may need to access knowledge from employees in other departments
to illuminate a set of alternative paths to solution, which is a likely contributor to the
perceived value of cross-departmental knowledge on divergent thinking. On the
other hand, individuals working on complex tasks may need to access knowledge
from employees in other departments to help evaluate potential solutions and hone
those that are most feasible, which is a likely contributor to the perceived value of
cross-departmental knowledge on convergent thinking.
3.6.3 People-to-people Interaction
People-to-people interaction describes the extent to which individuals use
voice-to-voice knowledge transfer channels to access knowledge from other people
(Alavi & Leidner, 2001). Most notably, these channels support socialization (Alavi
& Leidner, 2001). Socialization is used to transfer the deep knowledge of one
person to another person (Nonaka & Takeuchi, 1995). By providing access to rich
contextual details it facilitates the interpretation and adaptation of knowledge (e.g.
Dixon, 2000; Dyer & Noveoka, 2000; Hansen et al., 1999; Hargadon & Sutton,
1997; Kanter, 1988; Leonard & Sensiper, 1998; Majchrzak et al., 2004; Nahapiet &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 1
Ghoshal, 1998; O'Dell & Grayson, 1998). Research also suggests that people-to-
people interaction can support the wide-dissemination of knowledge, which is likely
to result in accessing knowledge that is more diverse and less redundant (Alavi &
Leidner, 2001). On the whole, people-to-people interaction is likely to contribute to
the perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking. With people-to-people interaction, individuals will not only
have access to diverse knowledge, but a means to understand how it might contribute
to their work. In addition, access to the contextual details that arise from people-to-
people interaction may also enable a more meticulous analysis and evaluation of
knowledge that facilitates the selection and implementation of workable solutions.
3.6.4 Perceived Ease of Use of Technology
The perceived ease of use of technology describes the degree to which
individuals believe using technology will be free of effort (Davis, 1989). Prior
research suggests that it is not enough for individuals to believe technology is useful;
they also have to believe technology is easy to use (Davis, 1989). In other words, it
is not enough for individuals to believe that the use of technology features can help
them bridge cross-departmental barriers. The individuals also have to believe that
the technology is easy to use. Supporting this assertion, Yandenbosch and Higgins
(1995) found perceived ease of use of technology was positively related to model
building and model maintenance, two modes of learning that are divergent and
convergent by nature. Similarly, in this dissertation study, it is expected that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 2
perceived ease of use of technology will be positively related to the perceived value
of cross-departmental knowledge on divergent thinking and convergent thinking.
3.7 Conceptual Model
Using the theory of thought worlds and the theory of hermeneutic inquiry as
discussed in this chapter, the conceptual model is shown in Figure 3.2.
Figure 3.2 Conceptual Model
HI
H2
Control Variables:
+ Perceived openness of communication
+ People to people interaction
+ Perceived task complexity
Use of technology features
that support cross-
departmental understanding
Perceived value of cross-
departmental knowledge on
convergent thinking
Perceived value of cross-
departmental knowledge on
divergent thinking
+ Perceived ease of use of technology
The model can be broadly described as follows. In order to bridge the
barriers encountered during cross-departmental knowledge access, individuals have
to access knowledge that is a) associated with the expert source, b) interpretable, and
c) adaptable. Accessing knowledge in this manner is expected to positively relate to
individuals’ perceived value of cross-departmental knowledge on innovation, the
major outcome under study. Recall from chapter 2 that this outcome is comprised of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 3
the perceived value of cross-departmental knowledge on divergent thinking and the
perceived value of cross-departmental knowledge on convergent thinking.
To the left is the major independent variable under study - the use of
technology features that support cross-departmental understanding. Using
technology in this manner is expected to bridge the barriers to cross-departmental
knowledge access and, thereby positively relate to the perceived value of cross-
departmental knowledge divergent thinking and convergent thinking. In the lower
left hand comer, there are four control variables: a) perceived openness of
communication channels, b) people-to-people interaction, c) perceived task
complexity, and d) perceived ease of use of technology. They too are expected to
positively relate to the perceived value of cross-departmental knowledge on
divergent thinking and convergent thinking. The two major hypotheses are as
follows.
Hypothesis 1: The use of technology features that support cross-
departmental understanding is positively related to the
perceived value of cross-departmental knowledge on divergent
thinking, after controlling for perceived openness of
communication, people-to-people interaction, perceived task
complexity, and perceived ease of use of technology.
Hypothesis 2: The use of technology features that support cross-
departmental understanding is positively related to the
perceived value of cross-departmental knowledge on
convergent thinking, after controlling for perceived openness of
communication, people-to-people interaction, perceived task
complexity, and perceived ease of use of technology.
3.8 Exploratory Question
It is reasonable to believe that the work environment within which the
knowledge is being accessed might influence the relationship between the use of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 4
technology features that support cross-departmental understanding and the perceived
value of cross-departmental knowledge on innovation. However, much of the
research on technology supported KM has not examined contextual influences on the
use of KM technologies (Becerra-Femandez & Sabherwal, 2001). Thus this
dissertation study considers the following exploratory question.
Is the relationship between the use of technology features that
support cross-departmental understanding and the perceived
value of cross-departmental knowledge on innovation
moderated by other factors such as perceived openness of
communication, people-to-people interaction, perceived task
complexity, or the perceived ease of use of technology?
3.8.1 Perceived Openness of Communication
Whether the perceived openness of communication moderates the major
relationships under study is not clear. Perceived openness of communication may
lead individuals to believe they can bridge the barriers to cross-departmental
knowledge access. However, the use of technology features that support cross-
departmental understanding provides a means to actually bridge these barriers. On
the one hand, perceived openness of communication may lead to more frequent use
of technology features that support cross-departmental understanding. As a result,
the use of technology features that support cross-departmental understanding may
have a greater influence on the perceived value of cross-departmental knowledge on
divergent thinking and convergent thinking. On the other hand, the availability of
technology features that support cross-departmental understanding may be enough
for individuals’ to perceive the value of cross-departmental knowledge on divergent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 5
thinking and convergent thinking. Whether perceived openness of communication is
high or low may not make much difference under such circumstances.
3.8.2 People-to-people Interaction
Similarly, it is not clear from prior research whether the use of technology
and people-to-people interaction represent an additive relationship or a multiplicative
one. On the one hand, researchers have suggested that greater benefits can be reaped
from the use of technology when it is coupled with people-to-people interaction
(Goodman & Darr, 1998; Nohria & Eccles, 1992). This implies that there may be an
interactive effect, where high rather than low people-to-people interaction increases
the perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking. On the other hand, researchers have suggested that these two
kinds of knowledge access may complement or compete with one another (Hansen et
al., 1999; Olivera, 2000), which implies an additive relationship rather than a
multiplicative one.
3.8.3 Perceived Task Complexity
Researchers have shown that the complexity of a task can place demands on
technology features making them more or less effective (Goodman & Darr, 1998).
Complex tasks that require frequent access to knowledge from other areas of
expertise may place greater demands on technology (e.g. Becerra-Femandez &
Sabherwal, 2001). The use of technology features that support cross-departmental
understanding may be perceived to have a greater impact on the perceived value of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 6
cross-departmental knowledge on divergent thinking and convergent thinking under
these conditions. This is in contrast to low complexity tasks. Low complexity tasks
tend not to require individuals to access knowledge from different areas of expertise
(Zigurs & Buckland, 1998). Thus, fewer demands may be placed on technology
features that support cross-departmental understanding, causing the impact on the
perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking to be lower.
3.8.4 Perceived Ease of Use of Technology
The ease of bridging the barriers encountered when accessing knowledge
from employees in other departments may have differential effects on the perceived
value of cross-departmental knowledge on divergent thinking and convergent
thinking. While research suggests that bridging the barriers positively affects the
perceived value of cross-departmental knowledge on innovation (Dougherty, 1992),
research also suggests easy access to knowledge also increases its value (Borgatti &
Cross, 2003). Thus, it is possible that an interactive effect between the use of
technology features that support cross-departmental understanding and the perceived
ease of use of technology may have a more profound effect on the perceived value of
cross-departmental knowledge on divergent thinking and convergent thinking, than
either would alone. Individuals using technology features that support cross-
departmental understanding via a technology that is perceived easy to use, may
believe that the knowledge is more easily accessed, and thus may value it more than
knowledge accessed via a technology that is not so perceived.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 7
3. 9 Summary
This dissertation study draws from the theory of thought worlds and the
theory of hermeneutic inquiry to propose a set of technology features that support
cross-departmental understanding. It is hypothesized that the use of technology
features that support cross-departmental understanding is positively related to
individuals’ perceived value of cross-departmental knowledge on divergent thinking
and individuals’ perceived value of cross-departmental knowledge on convergent
thinking. This occurs after controlling for other factors in the work environment,
such as perceived openness of communication, people-to-people interaction,
perceived task complexity, and perceived ease of use of technology. In addition, an
exploratory research question is posed that considers whether the four variables
being controlled act to moderate the two major relationships under study.
The next chapter describes the research setting at Company A. The
preliminary data collection activities are discussed as well. In addition, the two-
wave survey administration to scientists and engineers in Company A is described,
including the operationalization of the variables under study. The rationale behind
using hierarchical regression analysis as the primary data analytic technique is also
provided.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 8
CHAPTER 4: RESEARCH METHODS
To test the hypotheses, a two-wave field survey was conducted. Conducting
the research at an organization where cross-departmental knowledge access was
important to successfully completing tasks also helped capture the complexity of the
phenomenon. All surveys were obtained from a single site to control for the
availability of technology resources and the basic nature of how the work was
performed. Company A was chosen as the data collection site (section 4.1). To
ensure that the way employees at Company A accessed knowledge was accurately
incorporated into the survey, extensive interviews and observations needed to be
conducted. For one year prior to final survey administration, participant
observations of KM committee meetings, structured interviews with management,
and semi-structured case studies of employee problem solving incidents were used to
confirm the appropriateness of the constructs selected for the conceptual model and
contextualize the survey (section 4.2). Section 4.3 describes the two-wave survey
methodology, including the survey instruments employed. The use of hierarchical
regression analysis as the primary data analytic technique is also discussed (section
4.4).
4.1 Research Setting
Company A is a medium-sized (approximately 3,000 employees) scientific
and technical organization with the mission to help prevent and solve problems
associated with the design, deployment, and operation of large, highly complex
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 9
space systems. The majority of the scientific and technical staff at Company A
(66%) hold advanced degrees in the engineering and science disciplines. The field
survey was approved by senior management at Company A, and coordinated through
the KM office. Preliminary data collection activities at Company A showed that its
employees depended on knowledge from colleagues within the organization to bring
historical expertise to bear on the current work they were performing. Quite often
the knowledge employees depended on was in specializations that were different
from their own and the work they were performing required innovation. In addition,
employees at Company A had many KM technologies and features available to
access each other’s knowledge.
4.2 Preliminary Data Collection
Preliminary data collection methods included participant observations of KM
committee meetings (section 4.2.1), structured interviews with managers (section
4.2.2), and semi-structured case studies with employees (section 4.2.3). These
activities served several ends. They provided management’s perspective on the
needs and interests of employees accessing each other’s knowledge, including the
organizational and technological support available. The activities were also used to
verify how employees bridged cross-departmental barriers and confirm that
technology features existed to help bridge cross-departmental barriers.
Consideration was also given to other factors in the work environment that may have
bearing on the dependent variables and therefore should be controlled.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 0
4.2.1 Participant Observation of KM Committee Meetings
The KM committee consisted of corporate librarians, vice presidents, and
general managers. The committee was charged with devising future organizational
and technological plans that would address the ongoing needs of employees when it
came to accessing knowledge from colleagues within the organization. Data
collection began with participant observation of three KM committee meetings over
the course of several months. In addition to participant observation, KM committee
documentation was reviewed, and informal discussions with five members of the
committee were conducted. These data collection activities provided insight into the
KM committee’s view, regarding the needs and interests of Company A employees
accessing each other’s knowledge, during the course of their work.
For example, the KM committee acknowledged the need for the Company A
employee to bring the historical expertise of other employees to bear on current
problems. It is considered one of Company A ’s core competencies and one of the
main reasons that Company A remains competitive in its industry. Thus, one goal of
the KM committee was to provide continued support to employees such that they
believed there were benefits to reusing each other’s knowledge. The committee also
acknowledged that employees at Company A used an array of KM technologies to
access knowledge from each other. However, the committee also conceded that
knowing the type of KM technology being used was not enough to inform committee
decisions with respect to the future organizational and technological plans. The
committee also needed to have a better understanding of specific technology features
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
that facilitated employees to access each other’s knowledge. For example, the
committee wanted to understand how employees managed to access and reuse each
other’s knowledge when it is so deeply ingrained in context and laden with complex
rationale. The KM committee believed it understood how people-to-people
interaction was being used to address the challenge. However, understanding how
employees were using technology to do so, especially when the knowledge was
outside of their own department was also a goal.
4.2.2 Structured Interviews with Management
To learn more about the organizational and technological environment
believed to facilitate knowledge access within Company A, the KM committee
suggested several managers to interview. Structured interviews were done with five
general managers across five divisions. From the managers’ perspective, employees
accessed each other’s knowledge as part of their daily job. Given its usual
occurrence, it was not usually associated with special rewards or recognition. While
most managers believed that Company A ’s culture promoted a sense of open
communication among employees, some did mention that it may be less evident
across departments. Supportive KM technologies made available to employees were
also discussed. A lot of the KM technology support was homegrown within the
division, subdivision, or department. For example, source code libraries provided
access to software for reuse during modeling and simulation analyses. Providing
access to the actual source code enabled individuals to better interpret and adapt (i.e.
change and recompile) it to fit the needs of their current context. Various
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 2
documents, briefings, and memos that were produced by employees during the
course of their work were also catalogued and made available for reuse via shared
databases. These documents were purposefully created at various levels of detail to
support employees’ need to revisit the rationale behind past problem solutions.
Historical details associated with rocket and satellite launches were captured over
time in corporate repositories. Providing access to details over time helped
Company A employees interpret how situations unfolded and were adapted to
changing contexts. The employees also used their email accounts to store documents
from briefings and seminars they attended as well as past correspondence with other
Company A employees. These storage spaces provided access to knowledge derived
from the diverse set experts that were working at Company A.
Structured interviews were also conducted with three subdivision managers
and one design center manager. In these interviews, the managers were asked about
the kinds of problems employees were tasked to solve, the KM strategies employees
used, and the supportive KM technologies that were available. The managers named
many of the same KM technologies described above as well as others, including
homegrown software modeling tools, intranet web pages, shared file servers,
distribution lists, and instant messaging. A list of KM technologies available at
Company A are described in Table 4.1, including examples of their use and whether
they support access to knowledge that is a) associated with the expert source, b)
interpretable, c) adaptable. However, the managers did not believe that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
availability and use of KM technology precluded the use of people-to-people
interaction or what one person called the “sneaker net”.
Table 4.1 KM Technologies Available for Use at Company A_____________
IT Systems Examples of Use Examples of knowledge access
supported
Email Email and email attachments from
colleagues that have been filed in email
folders. Examples include briefings
(e.g. slide presentations) from technical
seminars or peer reviews and questions
asked and answered.
Authorship of emails and
attachments associates the
knowledge with the expert.
Shared Databases
Shared File Servers
Link to various documents. For
example Company A has a series of
memos that are categorized and named
based on content and audience. For
example, certain memo types are known
to provide vary detailed technical
information while others are known to
provide an overview o f a problem. One
memo type might describe how to
design a particular system by detailing
the components used for the design and
the associated parameters. Another
memo type might describe the resolution
of a failure or anomaly by: stating the
problem, reproducing the failure,
describing the steps used to implement
the solution including assumptions,
constraints, and rationale.
Knowledge at various levels of
detail supports interpretation.
Authored documents identify
expert sources.
Distribution Lists
List Servers
Distribute problem alert bulletins
describing problems with parts, material,
and processes.
Being alerted to changes made
to knowledge supports
adaptation.
Corporate Repositories Provide vast amount of historical details
on rocket and satellite launches
including performance characteristics,
failures, and configurations. Launch
vehicle, space program, flight number as
well as other characteristics are used to
categorize the details.
Tracing knowledge as it unfolds
over time supports adaptation.
Access to the historical details
facilitates interpretation.
Instant Messaging
Live Discussion Groups
Employees exchange messages in close
to real time.
The conversation style quality
supports interpretation and
adaptation of expert knowledge.
Experts are identified.
Source Code Libraries Provides software routines for reuse.
Some software routines are only
available in complied versions and have
to be reused as-is. Other software
routines make the source code available
to be changed and recompiled.
Accessing source code enables
access to programmer reasoning
and facilitates changing the
source to meet one’s own needs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The managers also provided insight into the kinds of work employees were
tasked to do. The problems Company A employees solved were characterized as
being fairly complex. As one manager put it, employees often encountered problems
that they could not solve with their own knowledge. They were expected to look at a
problem from different perspectives and come up with alternative approaches to
solve it. For example, one employee had to solve a satellite-processing problem.
The satellite had 7400 variables and several potential influencing factors, including
new test equipment, new cables, and new work standards. To solve the problem he
had to learn all of these new things. Faced with problems like this, employees at
Company A frequently accessed knowledge from colleagues in other departments
with different areas of expertise. According to the managers, when solving these
kinds of problems, the employees’ job was not to simply reapply past solutions they
may have seen or heard about from other people in the organization. Instead, they
were expected to constantly improve upon past solutions as well as minimize the
potential of future problems.
4.2.3 Semi-structured Case Studies
The case studies lent insight into how the employees actually went about
accessing and reusing each other’s knowledge during the course of their work,
including the technologies they employed. The case studies were conducted with
three objectives in mind. The first was to verify that employees bridged barriers by
accessing knowledge that was: a) associated with the expert source, b) interpretable,
c) adaptable. The second was to confirm that technology features existed to help
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 5
bridge barriers. The third was to consider whether the variables being controlled for
were appropriate. The three subdivision managers and the design center manager
were asked to nominate two employees for interviews and identify whether the
employees made innovative contributions or played a supporting role. In all eight
employees were interviewed: four were nominated from the subdivisions and four
were nominated from the design center. The focus of these semi-structured
interviews was on the employees telling stories about when they solved a problem by
reusing knowledge from other employees.
The findings from the case studied verified how cross-departmental barriers
were bridged (Table 4.2). Employees did access knowledge that was associated with
the expert source. It was critical that the employees located experts and tangible
representations of their expertise (Faniel & Majchrzak, 2003). To support access to
knowledge that was interpretable, employees relied on acquiring knowledge at
various levels of detail (Faniel & Majchrzak, 2003). To bridge cross-departmental
barriers, knowledge also had to be flexible enough to be adapted to meet employees’
current problem solving context (Faniel & Majchrzak, 2003).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
Table 4.2 Case Study Findings that Verify How Company A Employees Bridge
Cross-departmental Barriers
Findings indicated
employees:
Example Quotes
Access knowledge that
is associated w/expert
source.
“[I looked for Bill]. Bill designed power supplies 20 years ago. When we had a
[satellite processing] problem I remembered Bill had problems. I used to meet
him for lunch when he worked [here]. [Then I looked for and] found his notes in
a file cabinet with file folders that contain everything about power supplies [...]
the power grid had changed a bit so I plugged in a new power supply and found
[the solution to the problem].”
Interpret knowledge by
accessing the big
picture and the fine
grained details of
knowledge
“The general physics is there but application to our specific waveband is not there
[...], also the effects of winds, atmospheric chemistry [are not there, we need
them to play together [so that we can understand how all the different aspects of
knowledge work together].”
Need to flexibly
change the knowledge
to fit their needs
“What’s spectacular about [the model] is the layout, [I can model] six different
thrusters, with different propellant [options]. The model allows me to do a bunch
more calculations [based on my problem solving needs].
The findings from the case studies also confirmed that technology features
existed to help employees bridge cross-departmental barriers (Table 4.3). For
example the use of on-line organizational charts supported the principle of
ownership, by allowing employees to click on people’s names to see what kind of
work they did. In support of the principle of multiplicity, employees were found to
use email to gather multiple perspectives on how to solve a problem. Company A
employees often spoke about and represented knowledge in terms of components and
component relationships, which allowed them to deconstruct and reconstruct the
contextual details about the knowledge for improved understanding (Faniel &
Majchrzak, 2003). Case studies found that some KM technologies were designed to
enable easy travel among these elements of knowledge to help employees gain
insight into different areas of expertise. Many of the KM technologies also
supported the flexible adaptation of knowledge, including ongoing updates that
refine knowledge for increased accuracy.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 7
Table 4.3 Case Study Findings that Confirm Technology Features Exist to Help
Bridge Cross-departmental Barriers__________________________________
Technology Design
Principles and
Definitions
Example Quotes
Ownership identifies
experts of knowledge
based on existing skills
and experiences
“The company has functional and formal org charts where I could go [...] and
click on the organization and find names o f people and click on another chart to
see what people do.”
Multiplicity provides
multiple interpretations
of the same or similar
situations from
different points of
view
“I knew some other people in another organization [that] support satellites and
had used XML with telemetry and emailed them [...] to gather 3 to 4 examples
and wrote down criteria of what a good solution would be based on what had
been done previously.”
Easy travel enables
movement across
different levels of
detail within and
across knowledge
representations
“[There are] ten different types of hardware with many different vendors [and
their technical specs. The model also contains] calculations from parameters into
how the satellite behaves [when there are disturbances in space] which
determines the kind of hardware [that is needed].”
Indeterminacy leaves
knowledge
representations open to
interpretation
“I created the substance [of the model. However] whenever someone wants to
change it [to fit his needs] he does.”
Emergence creates and
refines knowledge
categories, constructs,
and levels of
abstraction
“I do [ongoing] updates to increase the accuracy of [the modeling] tool [for
example changes I made improved accuracy] from 60% to 80%.”
The case studies also confirmed that other factors in the work environment
should be controlled (Table 4.4). For example, as noted in the interviews with
management, findings from the case studies showed that employees used people-to-
people interaction and technology to access knowledge from each other (Faniel &
Majchrzak, 2002, 2003). Case study findings also showed employees may not be as
open to communication with colleagues in other departments as they are with
colleagues in their own department. The findings also indicated that the perceived
ease of use of technology and perceived task complexity should be controlled for in
this dissertation study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 8
Table 4.4 Case Study Findings that Confirm Other Factors in the Work Environment
Should be Controlled
Control Variables Example Quotes
Openness of
communication
“Some people have a not invented here philosophy. [This does not occur] in my
group, but in others if not developed [in their group it is] not good.”
Opportunity for people
to people interaction
“ [...] my boss put me on a program that had small interdisciplinary teams one of
each type of engineer [...] I built a network from there, if they did not know they
would [send me to someone].”
Task complexity “[The problem was a challenge] the satellite had 7400 variables, six monitors
were noisy [and there were] several potential influencing factors [including] new
test equipment, new cables, new work standards, new satellite. I had to learn
these new things.”
Ease of use of
technology
“People here resist using [that IT system], because [it is] difficult to use and
different.”
4.2.4 Summary
In sum, preliminary data collection activities included participant observation
of the KM committee, structured interviews with management, and semi-structured
case studies with scientist and engineers. These preliminary data collection activities
verified that Company A was a good testing ground for the hypotheses presented in
this study. Findings verified that employees at Company A bridge cross-
departmental barriers by accessing knowledge that is: a) associated with the expert
source, 2) interpretable, and 3) adaptable. Findings also confirmed that technology
features existed to help bridge cross-departmental barriers. In addition, the control
variables included in the conceptual model seemed to be appropriate given Company
A ’s work environment. The preliminary data collection activities also helped
contextualize the survey. For example, findings indicated that employees had to rely
on knowledge from colleagues in Company A with different areas of expertise.
Within Company A, areas of expertise tended to change at the department level. To
capture this, the survey items had respondents refer to the acquisition of knowledge
from Company A coworkers outside rather than within their own department.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 9
4.3 Survey Methodology
Prior to the formal administration of the two-wave survey, the full survey was
piloted five times with five Company A employees. For each pilot employees were
met with individually for 45-60 minutes and instructed to read the survey and answer
each question. Employees were also encouraged to question and comment on the
clarity, appropriateness, and understandability of survey instructions, questions, and
items. Based on feedback during each pilot, changes were made to the survey in
preparation for the next pilot. The process was repeated until the survey instructions,
questions, and items were considered to be clear, appropriate, and well understood.
Next, the formal two-wave survey was administered to test the major
hypotheses under study. The first wave survey was administered in the Fall 2003
and the second wave survey was administered in the Spring 2004 (section 4.3.1). All
the survey instruments were adapted from existing literature, except for two, which
were created based on existing literature and the preliminary data collection efforts
(section 4.3.2).
4.3.1 Survey Administration
Both the first wave survey and the second wave survey were administered via
the World Wide Web and hosted by a third party web based survey company
(section 4.3.1.1). Power analysis indicated that the response rates for the first and
second wave surveys provided large enough sample sizes to detect a medium effect
size during hierarchical regression analysis (section 4.3.1.2). A comparison of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 0
characteristics between respondents and non-respondents showed that non-response
bias was not a problem (section 4.3.1.3).
4.3.1.1 Two-Wave Survey
A two-wave survey design was used to minimize common method bias
(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The first wave survey collected
data for both independent and dependent variables in the conceptual model -
perceived openness of communication, people-to-people interaction, perceived task
complexity, perceived ease of use of technology, the use of technology features that
support cross-departmental understanding, perceived value of cross-departmental
knowledge on innovation (i.e. divergent thinking and convergent thinking)
(Appendix A). The second wave survey collected dependent variable data only -
perceived value of cross-departmental knowledge on innovation (i.e. divergent
thinking and convergent thinking) (Appendix B).
The sampling pool for the first wave survey consisted of 874 scientists and
engineers across five divisions. Potential respondents were sent an email invitation
endorsed by the vice president of the five divisions in the Fall 2003. A week later
those that had not submitted a survey were sent a reminder. Two weeks later those
that had not submitted a survey were sent a second reminder. All three emails
contained a hyperlink to the survey web page. The emails and the survey had written
text informing potential respondents that the surveys were voluntary. Since the
respondents’ badge numbers were used to identify them for the second wave survey,
both the email and the online survey assured potential respondents that their
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
identities would be kept confidential. Of the 874 scientists and engineers invited to
complete the survey, 310 responded (35.5%). Of the 310 surveys that were
submitted, 248 (28.4%) were available for data analysis. One survey was deleted
because the respondent recorded the same response for every question. The others
had missing data problems.
The sampling pool for the second wave survey consisted of the 248
respondents from the first wave survey. The 248 potential respondents were sent an
email invitation in the Spring 2004. Similar to the first wave, two email reminders
were sent a week apart, all three emails contained a hyperlink to the survey web
page, the emails and the survey had written text informing potential respondents that
the surveys were voluntary and that their identities would be kept confidential. Of
the 248 invited to complete the survey, 131 (52.8%) were returned, complete, and
available for data analysis.
4.3.1.2 Power Analysis
Cohen’s (1988) power analysis equations were used. A power of .80, and a
significance criterion of a = .05 were given. Taking the number of independent
variables in the conceptual model into consideration, the final sample sizes for the
first (N=248) and second (N=131) waves were large enough to detect a medium
effect size during hierarchical regression analysis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 2
4.3.1.3 Assessing Non-response Bias
The first and second waves of data were tested for non-response bias in two
ways. First, two characteristics known about the sampling frame were compared for
respondents and non-respondents (Mayer & Pratt, 1966). For the first and second
wave surveys, tenure and division were recorded for each respondent group and the
total sample in Table 4.5 and Table 4.6 respectively.
Table 4.5 First Wave Characteristics for Response Groups and Total Sample
Total Total Total
Respondents Non-respondents Sample Population
Tenure
Greater than 15 years 96 (38.7%) 265 (42.3%) 361 (41.3%)
Greater than 6 years 63 (25.4%) 173 (27.6%) 236 (27%)
0 to 5 years 89 (35.9%) 188 (30%) 277 (31.7%)
Total 248 626 874
Division Group
Division C 49 (19.8%) 111 (17.7%) 160(18.3%)
Division E 50 (20.2%) 134 (21.4%) 184(21.1%)
Division L 43 (17.3%) 124 (19.8%) 167 (19.1%)
Division S 50 (20.2%) 112(17.9%) 162 (18.5%)
Division V 56 (22.6%) 145 (23.2%) 201 (23%)
Total 248 626 874
Table 4.6 Second Wave Characteristics b r Response Groups and Total Sample
Total Total Total
Respondents Non-respondents Sample Population
Tenure
Greater than 15 years 48 (36.6%) 48 (41.0%) 96 (38.7%)
Greater than 6 years 35 (26.7%) 28 (23.9%) 63 (25.4%)
0 to 5 years 48 (36.6%) 41 (35.0%) 89 (35.9%)
Total 131 117 248
Division Group
Division C 27 (20.6%) 22 (18.8%) 49 (19.8%)
Division E 25(19.1% ) 25 (21.4%) 50 (20.2%)
Division L 22 (16.8%) 21 (17.9%) 43 (17.3%)
Division S 25 (19.1%) 25 (21.4%) 50 (20.2%)
Division V 32 (24.4%) 24 (20.5%) 56 (22.6%)
Total 131 117 248
Badge numbers for all employees at Company A were known. Company A issues
the badge numbers serially and the numbers can be used to represent tenure at
Company A. The division assignment for all employees at Company A was also
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 3
known. In both cases, a chi-square test indicated that there was no significant
difference across groups.
Next variable means of the first wave data were compared (Filion, 1975-
1976). To rule out non-response bias for the first wave, the variable means were
compared for early and late respondents. Late respondents are expected to be most
similar to non-respondents (Filion, 1975-1976; Mayer & Pratt, 1966). A one-way
ANOVA test showed early responders to the email invite were not significantly
different than late responders to the first or second email reminders (Table 4.7).
Table 4.7 First Wave Variable Means over 1'hree Weeks of Response
Survey
Invitation
First
Follow-up
Second
Follow-up
(N = 145) (N = 69) (N = 34)
Perceived Value of Cross-departmental Knowledge on
Divergent Thinking
2.74 2.67 2.75
Perceived Value o f Cross-departmental Knowledge on
Convergent Thinking
2.46 2.39 2.51
Perceived Openness o f Communication 3.86 3.94 3.98
Perceived Task Complexity 4.09 3.99 4.04
Corporate Initiated Encounters 2.22 2.16 2.32
Personally Initiated Encounters 3.08 2.83 3.01
Perceived ease o f use of technology 3.19 3.29 3.51
Use of Technology Features that Support Cross-
departmental Understanding
2.55 2.39 2.76
To rule out non-response bias for the second wave, the variable means from
first wave data were compared. This time, variable means for respondents who
completed the first, but not the second wave survey were compared with respondents
who completed both the first and second wave surveys. A one-way ANOVA test
showed no significant difference between the two groups (Table 4.8).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 4
Table 4.8 First Wave Variable Means for Respondents and Non-respondents of the
Second Wave
Respondents Non-respondents
(N = 131 ) (N = 117)
Perceived Value of Cross-departmental Knowledge on
Divergent Thinking
2.78 2.66
Perceived Value of Cross-departmental Knowledge on
Convergent Thinking
2.44 2.45
Perceived Openness of Communication 3.93 3.87
Perceived Task Complexity 4.13 3.97
Corporate Initiated Encounters 2.22 2.22
Personally Initiated Encounters 3.00 3.01
Perceived ease of use of technology 3.36 3.15
Use of Technology Features that Support Cross-
departmental Understanding
2.55 2.51
4.3.1.4 Summary
A two-wave survey methodology was employed. The first wave survey
collected data on independent and dependent variables, whereas the second wave
survey collected data on the dependent variables only. In the first wave 248 out of
874 surveys were returned and used in data analysis (28.4%). In the second wave
131 of the 248 surveys were returned and used in data analysis (52.8%). The second
wave data only represents 15% of the 874 originally surveyed. While a 15%
response rate may be considered low, in absence of non-response bias, it does not
pose a serious limitation. Moreover, the power analysis indicates that the sample
size is adequate to detect a medium effect size.
4.3.2 Survey Instruments
Two survey instruments - the use of technology features that support cross-
departmental understanding (section 4.3.2) and people-to-people interaction (section
4.3.3) were created. The remaining survey instruments, which were adapted from
existing literature include: perceived openness of communication (section 4.3.4),
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 5
perceived task complexity (section 4.3.5), perceived ease of use of technology
(section 4.3.6), and perceived value of cross-departmental knowledge on innovation
(i.e. divergent thinking and convergent thinking) (section 4.3.7).
4.3.2.1 Use of Technology Features that Support Cross-departmental Understanding
Respondents began the survey by indicating the frequency they used a list of
ten technologies available at Company A (Table 4.9). The list of ten information
technology (IT) systems was created based on the preliminary data collection at
Company A. The primary purpose of this scale was to create a common context of
what was meant by IT systems. Given that preliminary data collection indicated that
the employees at Company A used a wide array of KM technologies, it was not
expected that this scale would create a valid, reliable measure of general IT systems
use that could be used during hierarchical regression analysis. This expectation was
confirmed.
Table 4.9 Items for IT Systems Available for Use at Company A
Question Stem: In the course of doing your regular professional work
at Company A, you access knowledge from Company A coworkers,
please use the scale below to indicate the frequency you use the
following IT systems to access the knowledge._______________________
Email (EMAIL)__________________________________________________
Databases (DB)__________________________________________________
Intranet (INTRANET)____________________________________________
Instant messaging, live discussion groups (IM)______________________
Modeling/simulation tools (MODELS)_____________________________
Shared file servers (FILESRYS)___________________________________
Knowledge management infrastructure (KMI)_______________________
Threaded discussions, electronic bulletin boards (ELECBB)___________
Sources code libraries/databases (CODELIBS)______________________
Distribution lists, list servers (DLIST)______________________________
Scale: l=none of the time, 2=some of the time, and 3=most of the time
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
Table 4.10 Items and Factor Analyses for the Use of Technology Features that
Support Cross-departmental Understanding___________ ______________
Technology Design
Principle Definitions
Question Stem: In the course of doing your regular
professional work at Company A, you access knowledge
from Company A coworkers in other departments/areas,
indicate how frequently the IT systems you indicated
using are used to...
First
Wave
Factor
Loadings
(N=248)
Second
Wave
Factor
Loadings
(N=131)
Ownership identifies
experts of
knowledge based on
existing skills and
experiences.
identify the experts (e.g. authors) of the knowledge (e.g.
memos, source code). (OWNR1E)
.680 .661
find others with the experience you need (e.g. specific
skills, program insights). (OWNR2E)
.705 .746
find knowledge that has been contributed by specific
individuals. (OWNR3E)
.765 .725
Multiplicity
provides multiple
interpretations of the
same or similar
situations from
different points of
view.
identify multiple ways to approach your problem/task
(e.g. competing theories, designs, algorithms).
(MULT IE)
.788 .762
access knowledge that offers different points of view
(e.g. comparisons between parts, materials, or designs).
(MULT2E)
.807 .783
find alternative ideas (e.g. compare output from or
perform what if scenarios). (MULT3E)
.765 .735
Easy travel supports
movement across
different levels of
detail within and
across
representations.
find additional knowledge that shares a similar subject
or purpose (e.g. sources referenced in memos, reports,
briefings). (TRAV1E)
.764 .758
access details that put the knowledge into context (e.g.
technical specifications, input parameters). (TRAV2E)
.759 .758
find summaries (e.g. technical narratives, design
reviews) as well as details (e.g. test data, source code,
technical configurations) about the knowledge.
(TRAV3E)
.785 .783
Indeterminacy
leaves
representations open
to interpretation.
access suggested solutions, thoughts, opinions, and/or
analyses (e.g. future predictions on part reliability, trend
analyses). (INDETIE)
.740 .753
find knowledge that supports the rationale behind the
decisions made by other people so that the decisions and
rationale can be revisited later (e.g. documented
assumptions, constraints, decision trees). (INDET2E)
.796 .809
find associated material that supports the knowledge
(e.g. notes, email, raw data, or schematics). (INDET3E)
.729 .743
Emergence creates
and refines
knowledge
categories and
constructs and levels
of abstraction.
find the changes made to the knowledge as ideas evolve
over time (e.g. access current and historical program
data). (EMRG1E)
.681 .695
keep track of how the knowledge changes over time
(e.g. account for design, material, or part changes over
life of a program). (EMRG2E)
.694 .708
be informed when knowledge you have an interest in
changes (e.g. reports of latest failures/anomalies).
(EMRG3E)
.730 .727
a = .94 a = .94
Scale: l=to no extent, 2=to a minor extent, 3=to some extent, 4=to a major extent, 5=to a great extent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 7
Next, based on the definitions of the five technology design principles -
ownership, multiplicity, easy travel, indeterminacy, emergence - a fifteen-item scale
was created to measure the use of technology features that support cross-
departmental understanding (Table 4.10, Columns 1 and 2). The initial scale was
created primarily from the work of Boland et al. (1994). Supporting IS literature
(Boland & Tenkasi 1995; Hirschheim & Klein, 1994; Markus et al., 2002; Mason &
Mitroff, 1973) was also used. The principle of ownership identifies experts of the
knowledge. The items reflect the need to find knowledge associated with known
experts as well as find experts based on the kind of knowledge being sought.
Multiplicity provides multiple points of view on situations. The items reflect the
need to identify various different approaches to work. Easy travel facilitates
movement among different layers of context and the items reflect the need to manage
access to knowledge at different levels of detail. Indeterminacy supports
impermanence and the items reflect the need to reinterpret existing knowledge in a
new light. Emergence supports change. The items reflect the need be kept abreast of
knowledge as it unfolds.
The initial scale was refined during three survey pilots conducted over the
course of six months. The first survey pilot was with a class of MBA students
(N=21). Three items were associated with each design principle. Based on the
results of a principal component factor analysis with varimax rotation, survey items
were refined in preparation for the second pilot. The second survey pilot was with a
class of undergraduate students (N=27). Four items were associated with each
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
design principle in the event that items had to be dropped during factor analysis. In
preparation for the third pilot, the items were refined based on the results of a
principle component factor analysis with varimax rotation. The third survey pilot
was with a group of practitioners (N=15). Again, four items were associated with
each principle in the event that items had to be dropped during factor analysis. The
final survey instrument retained three items for each design principle. The
preliminary data collection at Company A helped to further contextualize the survey
instrument, including the addition of Company A specific examples to the items.
The results of factor analysis on the first and second waves of data are shown
in Table 4.10, Columns 3 and 4. The factor loadings across both waves were
acceptable. In both cases, the total variance explained for the first and second wave
factors was 56%. The Cronbach’s alphas were also acceptable at .94 for each wave.
4.3.2.2 Perceived Openness of Communication
A five-item scale was adapted from O’Reilly and Roberts (1977) to measure
perceived openness of communication. The O’Reilly and Roberts (1977) scale was
worded to measure perceived openness of communication within a group (Table
4.11, Column 1). In this dissertation study, the items were reworded to measure
perceived openness of communication among employees at Company A in general
(Table 4.11, Column 3).
One factor was expected and shown in the first and second wave factor
analyses (Table 4.11, Columns 4 and 5). The results compared favorably to those in
the O ’Reilly and Roberts’ (1977) study (Table 4.11, Column 2). The total variance
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 9
explained in the first and second wave factor analyses was 67% and 62%
respectively. The Cronbach’s alphas for the first and second waves were adequate at
.87 and .84 respectively.
Table 4.11 A Comparison of Items and Factor Analyses for Perceived Openness of
Communication: O ’Reilly and Roberts vs. Dissertation Study_________
O’Reilly and Roberts Dissertation Study
Items Factor
Loadings
Question Stem: Please
indicate the extent you agree
with the following statements
about communication between
you and Company A
coworkers in other
departments/areas.
First
Wave
Factor
Loadings
(N=248)
Second
Wave
Factor
Loadings
(N=131)
It is easy to talk openly to all
members of this group.
.71 It is easy to talk openly to
employees at Company A
(COMMIE).
.835 .826
Communication in this group
is very open.
.72 Communication at Company
A is very open. (COMM2E)
.832 .767
I find it enjoyable to talk to
other members of this group.
.75 I find it enjoyable to talk to
others at Company A.
(COMM3E)
.791 .759
When people talk to each
other in this group, there is a
great deal of understanding.
.74 When people talk to each
other there is a great deal of
understanding. (COMM4E)
.775 .741
It is easy to ask advice from
any member of this group.
.75 It is easy to ask advice from
any Company A employee.
(COMM5E)
.845 .837
fi
II
00
00
a = .87
00
II
3
Dissertation Scale: l=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
4.3.2.3 People-to-people Interaction
A ten-item scale was used to measure people-to-people interaction. The
initial list of items was based on prior research (Alavi & Leidner, 2001; Becerra-
Femandez & Sabherwal, 2001; Goodman & Darr, 1998). For example, unscheduled
meetings, informal seminars, training, and personnel transfer, projects, corporate
meetings, and phone calls, have been suggested as a means of people-to-people
interaction. The set of existing items was refined, and new items were added during
preliminary data collection at Company A. For example, the training item was
replaced with the Company A Institute courses item, the informal seminars item was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 0
replaced with the briefings item. Employees at Company A also used additional
kinds of people-to-people interaction including: informal peer reviews, guest
lectures, technical seminars, and corporate events. All of these were added the
survey instrument (Table 4.12, Column 1).
Table 4.12 Items and Factor Analyses for People-to-people Interaction
Question Stem: In the course of doing
your regular professional work at
Company A, you access knowledge from
Company A coworkers in other
departments/areas, please use the scales
below to indicate how frequently you use
the following activities to access the
knowledge.
First Wave (N=248) Second Wave(N=131)
Factor 1
Loadings
Factor 2
Loadings
Factor 1
Loadings
Factor 2
Loadings
Corporate Initiated Encounters
Guest lectures (GLECTE) .816 .108 .837 .096
Company A Institute courses (AICE) .775 .075 .731 .049
Corporate events (e.g. CEO’s report to
employees) (CORPE)
.721 .028 .648 .191
Technical seminars (TSEME) .663 .288 .688 .122
Personally Initiated Encounters
Common projects worked on in the past
(e.g. you and co-worker(s) assigned to
same project) (PROJE)
.041 .801 .061 .765
Unscheduled meetings (e.g. cafeteria,
hallway, desk) (UMTGE)
.099 .694 .124 .646
Telephone conversations (TELE) .138 .711 .143 .681
Briefings (e.g. verbal presentations)
(BRIEFE)
.198 .725 .316 .680
Informal peer reviews (IPRE) .069 .704 .005 .742
Video conferences (VCONFE) Deleted Deleted Deleted Deleted
a = .75
O N
It
8
a = .7 2 a = .76
Scale: l=to no extent, 2=to a minor extent, 3=to some extent, 4=to a major extent, 5=to a great extent
The results from factor analysis on the first and second waves of data are
shown in Table 4.12, Columns 2-5. A single factor was expected, but results for
both waves showed a two-factor solution. The first factor, corporate initiated
encounters, included: technical seminars, corporate events, Company A Institute
courses, and guest lectures. The second factor, personally initiated encounters,
included: common projects worked on in the past, briefings, informal peer reviews,
unscheduled meetings, and telephone conversations. One item (video conferences)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
was deleted for low factor loadings during both factor analyses. The two-factor
solution explained 56% of the total variance for the first wave and 53% of the total
variance for the second wave. Both factors were shown to be reliable for both waves
of data analysis, with Cronbach’s alphas above .70 (Nunnally, 1978).
4.3.2.4 Perceived Task Complexity
A fourteen-item scale developed by George and Zhou (2001) was used to
measure perceived task complexity. No changes were made to the wording of the
items. According to the authors individuals can perceive tasks to be complex for
three reasons. First, the means to completing the task may be unclear. Second, the
ends (e.g. goals, objectives) of the task may be unclear. Third, there may be multiple
ways to go about completing the task. The authors treated these as three different
measures of task complexity.
A three-factor solution was expected. However, several items had poor
factor loadings and were deleted during the first and second wave factor analyses.
Unlike George and Zhou’s (2001) findings, the first and second wave factor analyses
revealed that respondents did not distinguish between tasks that had unclear means
vs. unclear ends. As a result, these two measures of perceived task complexity were
not used in this dissertation study. Similar to George and Zhou’s (2001) findings,
the first and second wave analyses confirmed a factor for tasks described as having
multiple paths to solution. This measure of perceived task complexity was retained
(Table 4.13). The total variance explained for the first and second wave factor
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 2
analyses was 79% and 75% respectively. The Cronbach’s alphas for the first and
second wave analyses were acceptable at .87 and .83 respectively.
Table 4.13 Items and Factor Analyses for Perceived Task Com]plexity
Question Stem: In the course of doing your regular
professional work at Company A, please indicate the extent
you agree with the following statements.
First Wave
Factor Loadings
(N=248)
Second Wave
Factor Loadings
(N=131)
Often, there are several different ways in which to perform
this job. (MMEANS1)
.870 .806
Many of the tasks on this job can be performed in different
ways. (MMEANS2)
.892 .886
More often than not, there are multiple ways to achieve work
goals or objectives. (MMEANS3)
.904 .898
Deciding which way to proceed on a work task or project is
often a challenge. (MMEANS4)
Deleted Deleted
There is often one best way to perform this job. (reverse
scored) (RMMEANS5)
Deleted Deleted
a = .87 a = .83
Scale: l=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
4.3.2.5 Perceived Ease of Use of Technology
A six-item scale was adapted from Davis (1989) to measure perceived ease of
use of technology (Davis, 1989). The Davis (1989) scale (Table 4.14, Column 1)
measured the perceived ease of use of a particular technology (i.e. CHART-
MASTER). The items used in this dissertation study were reworded to measure
perceived ease of use of technology for IT systems in general (Table 4.14, Column
3).
One factor was expected and shown in the first and second wave factor
analyses (Table 4.14, Column 4 and 5). The results compared favorably to those in
the Davis study (Table 4.14, Column 2). The total variance explained in the first and
second wave factor analyses was 75% and 71% respectively. The Cronbach’s alphas
for the first and second waves were acceptable at .93 and .91 respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 3
Table 4.14 A Comparison of Items and Factor Analyses for Perceived Ease of Use of
Technology: Davis vs. Dissertation Study____________________________
Davis Dissertation Study
Items Factor
Loadings
Question Stem: Please use the
scale below to indicate the
extent you agree with the
following statements about the
IT Systems you use when
accessing knowledge from your
coworkers at Company A.
First
Wave
Factor
Loadings
(N=248)
Second
Wave
Factor
Loadings
(N=131)
Learning to operate CHART-
MASTER would be easy for me.
.97 Learning to operate the IT
systems is easy for me. (EOU1)
.858 .804
I would find it easy to get
CHART-MASTER to do what I
want it to do.
.83 I find it easy to get the IT
systems to do what I want them
to do. (EOU2)
.915 .890
My interaction with CHART-
MASTER would be clear and
understandable.
.89 My interaction with the IT
systems is clear and
understandable. (EOU3)
.894 .875
I would find CHART-MASTER
to be flexible to interact with.
.63 I find the IT systems flexible to
interact with. (EOU4)
.802 .754
It would be easy for me to
become skillful at using
CHART-MASTER.
.91 It was easy for me to become
skillful at using the IT systems.
(EOU5)
.838 .847
I would find CHART-MASTER
easy to use.
.91 I find the IT systems easy to use.
(EOU6)
.897 .862
a = .91 a = .93 ct = .91
Dissertation Scale: l=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
4.3.2.6 Perceived Value of Cross-departmental Knowledge on Innovation
The scale for perceived value of cross-departmental knowledge on innovation
was adapted from Jabri (1991). The author used the scale to measure respondents’
preferred mode of thinking. Jabri (1991) asked respondents to indicate their answers
to the nineteen statements on a seven-point scale ranging from “unlikely to enjoy” to
“likely to enjoy” (Table 4.15, Column 1). Respondents in this dissertation study
were asked assess how accessing coworker knowledge from other departments
helped (Table 4.15, Column 4). Some items were reworded based on feedback from
the five Company A survey pilots.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 4
As expected the results of the factor analyses for the first and second waves
showed a two-factor solution, the perceived value of cross-departmental knowledge
on divergent thinking and the perceived value of cross-departmental knowledge on
convergent thinking. During the first wave factor analysis two items from each
factor, four items in all, were deleted because of low factor loadings (Table 4.15,
Columns 5 and 6). The resulting variance explained was 72%. These four items
were retained in the second wave (Table 4.15, Columns 7 and 8) and the variance
explained was 79%. The two factor solution for both the first and second wave
analyses compared favorably to the results from Jabri’s (1991) study (Table 4.15,
Columns 2 and 3). The Cronbach’s alphas for both factors across both waves of data
analysis were acceptable.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 4.15 A Comparison of Items and Factor Analyses for Perceived Value of Cross-departmental Knowledge on Innovation:
Jabri vs. Dissertation Study_________________________________________________________________________________
Jabri Dissertation Study
Item Factor 1
Loadings
Factor 2
Loadings
Question Stem: In the course of doing
your regular professional work at
Company A, indicate the extent accessing
knowledge from Company A coworkers in
other departments/areas helps you...
First Wave (N=248) Second Wave (N=131)
Factor 1
Loadings
Factor 2
Loadings
Factor 1
Loadings
Factor 2
Loadings
Being confronted with a maze of ideas
which may, or may not lead me
somewhere.
-.07 .58 Consider various problem solving ideas
that may or may not lead you to a solution.
(DIV1E/SDIV IE)
Deleted Deleted .412 .741
Pursuing a problem, particularly if it takes
me into areas I don’t know much about.
-.03 .56 Purse a problem, particularly if it takes
you into areas you don’t know much
about. (DIV2E/SDIV2E)
Deleted Deleted .229 .803
Searching for novel approaches not
required at the time.
-.09 .62 Search for novel approaches not required
at the time. (DIV3E/SDIV3E)
.286 .738 .352 .798
Linking ideas which stem from more than
one area of investigation.
.03 .61 Link ideas which stem from more than one
area of investigation. (DIV4E/SDIV4E)
.287 .809 .319 .842
Making unusual connections about ideas
even if they are trivial.
.06 .60 Make unusual connections about ideas
even if they are trivial. (DIV5E/SDIV5E)
.352 .768 .345 .827
Spending time tracing relationships
between disparate areas of work.
.01 .64 Trace relationships between disparate
areas of work. (DIV6E/SDIV6E)
.270 .799 .342 .809
Being fully occupied with what appear to
be novel methods o f solution.
-.04 .56 Engage in what appear to be novel
methods o f solution. (DIV7E/SDrV7E)
.448 .744 .376 .817
Struggling to make connections between
apparently unrelated ideas.
-.11 .66 Make connections between apparently
unrelated ideas. (DIV8E/SDIV8E)
.313 .834 .343 .839
Being ‘caught up’ by more than one
concept, method or solution.
-.13 .61 Think about more than one concept,
method or solution. (DrV9E/SDIV9E)
.472 .713 .367 .818
o
00
I t
S
a = .93 a = .96
Dissertation Scale: l=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
< 1
U i
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 4.15 continued
Jabri Dissertation Study
Item Factor 1
Loadings
Factor 2
Loadings
Question Stem: In the course o f doing
your regular professional work at
Company A, indicate the extent accessing
knowledge from Company A coworkers in
other departments/areas helps you...
First Wave (N=248) Second Wave (N=131)
Factor 1
Loadings
Factor 2
Loadings
Factor 1
Loadings
Factor 2
Loadings
Adhering to the commonly established
rules of my work area.
.63 -.15 Adhere to commonly established rules of
your area o f work. (CONV1E/SCONV1E)
Deleted Deleted .760 .318
Following well-trodden was and generally
accepted methods for solving problems.
.62 -.32 Follow generally accepted methods of
solving problems. (CONV2E/SCONV2E)
Deleted Deleted .739 .463
Paying strict regard to the sequence of
steps needed for the completion of a job.
.67 .02 Pay strict regard to the sequence of steps
needed for the completion of a job.
(CONV3E/SCONV3E)
.813 .281 .881 .287
Adhering to the well-known techniques,
methods, and procedures of my area of
work.
.76 -.17 Adhere to the well-known problem
solving techniques methods, and
procedures of your area of work.
(CONV4E/SCONV4E)
.813 .279 .860 .308
Accepting readily the usual and generally
proven methods o f solution.
.67 -.17 Readily accept the usual and generally
proven methods of solution.
(CONV 5E/SCONV5E)
.730 .239 .802 .371
Being precise and exact about production
of results and reports.
.69 .08 Be precise and exact about production of
results. (CONV6E/SCONV6E)
.777 .302 .787 .415
Adhering carefully to the standards of my
area of work.
.63 .07 Adhere carefully to the standards o f your
area of work. (CONV7E/SCONV7E)
.765 .415 .897 .273
Being fidly aware beforehand of the
sequence of steps required in solving
problems.
.59 -.14 Be fully aware o f the sequence of steps
required in solving problems.
(CONV8E/SCONV8E)
.814 .387 .823 .325
Being strict on the production o f results
as, and when required.
.60 .10 Be rigorous in deriving your solutions.
(CONV9E/SCONV9E)
.714 .420 .779 .420
Being methodical and consistent in the
way I tackle problems.
.66 -.01 Be methodical in the way you tackle the
problems. (CONVIOE/SCONVIOE)
.725 .424 .829 .359
a = .87 8
I I
a = .97
Dissertation Scale: l=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree
- J
ON
7 7
4.4 Data Analytic Techniques
The primary objective of this study is theory testing. The goal is to test direct
relationships between the independent and dependent variables. Of particular
interest is testing whether a positive relationship exists between the use of
technology features that support cross-departmental understanding and the perceived
value of cross-departmental knowledge on innovation. More specifically, it is
expected that the former will positively relate to the perceived value of cross-
departmental knowledge on divergent thinking and the perceived value of cross-
departmental knowledge on convergent thinking after controlling for other factors
including - perceived openness of communication, people-to-people interaction,
perceived task complexity, and the perceived ease of use of technology. Assessing
the additional variance explained by the use of technology features that support
cross-departmental understanding variable is also of interest. A secondary objective
of this dissertation study is to consider whether the five variables that are being
controlled interact with the use of technology features that support cross-
departmental understanding to strengthen individuals’ perceived value of cross-
departmental knowledge on divergent thinking and/or their perceived value of cross-
departmental knowledge on convergent thinking.
Based on the objectives of this dissertation and the sample sizes for the first
and second waves, hierarchical regression analysis was used as the primary data
analysis technique. First, the sample size for the first wave (N=248) and the second
wave (N=131) accommodated the use of hierarchical regression analysis without
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 8
undue loss of power. Using the same data analytic technique for both the first and
second waves of data analysis provided a basis to compare the results across both
waves. Third, the technique provides a relatively straightforward test of interaction
effects. The results of the hierarchical regression analyses done for the first and
second waves of data are reported in the next chapter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 9
CHAPTER 5: RESULTS
This chapter begins with the results from the first wave data analysis (section
5.1). Although hypotheses 1 and 2 were supported, two major limitations
compromised these findings - common method variance and discriminant validity.
The second survey, which collected dependent variable data only was administered
to address these limitations. Second wave data analysis also indicated that
hypotheses 1 and 2 were supported (section 5.2). The use of technology features that
support cross-departmental understanding was positively related to the perceived
value of cross-departmental knowledge on divergent thinking (hypothesis 1) as well
as the perceived value of cross-departmental knowledge on convergent thinking
(hypothesis 2). In answer to the exploratory question posed in chapter 3, the second
wave data analysis found neither one of the major relationships under study to be
moderated by other factors in the work environment (section 3.3).
5.1 First Wave Data Analysis
The results of tests of discriminant validity among constructs will be
discussed first (section 5.1.1). Next the descriptive statistics and correlations of first
wave variables will be presented (section 5.1.2), along with the findings from a
preliminary examination of the data to identify univariate outliers and evaluate
linearity and normality (section 5.1.3). Next, the results of hierarchical regression
analysis will be highlighted, including an evaluation of hierarchical regression
assumptions (section 5.1.4). This will be followed by a discussion of two major
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 0
limitations encountered during analysis of the self-reported, cross-sectional survey
data from the first wave - common method variance and discriminant validity
(section 5.1.5).
5.1.1 First Wave Tests for Discriminant Validity
The discriminant validity of constructs must be shown in two ways (Chin,
1998b). For the first test of discriminant validity, all items for all constructs were
subjected to a principal components factor analysis with varimax rotation. For
discriminant validity to be supported items must load higher on their own construct
than other constructs (Chin, 1998b). Items that cross-loaded were eliminated. Items
that had low loadings (below .50) on their own construct or loaded on another
construct at or above .50 were considered for elimination. During this first test of
discriminant validity, two items (EMRG1E, EMRG2E) from the use of technology
features that support cross-departmental understanding construct were eliminated for
cross loading on another factor. After the elimination of these two items, results
showed all remaining items loaded higher on their own construct than on any other
(Table 5.1).
Two items from the perceived value of cross-departmental knowledge on
divergent thinking construct (DIV7E, DIV9E) had loadings on the perceived value of
cross-departmental knowledge on convergent thinking construct of approximately
.50. However, the items were retained for several reasons. First the items were
adapted from an existing scale that has proven to be valid and reliable (e.g. Jabri,
1991; Scott & Bruce, 1994). In addition, there was no indication that the items were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
poorly worded. Third, recall from chapter 4 that confirmatory factor analysis of the
perceived value of cross-departmental knowledge on divergent thinking and
convergent thinking items by themselves showed convergent and discriminant
validity. Lastly, the results upon the elimination of these two items showed no
substantial improvement.
Table 5.1 First Wave Factor Solution with Varimax Rotation
1 2 3 4 5 6 7 8
TRAV1E .766 .092 .088 .062 .091 .020 -.014 .130
MULT IE .752 .235 .079 .080 .042 .101 .093 .015
OWNR3E .746 .197 .003 .065 .006 .151 -.135 .095
MULT2E .745 .273 .123 .086 .058 .134 .067 .004
MULT3E .735 .223 .100 .019 .042 .128 .019 -.070
INDET2E .726 .208 .094 .234 .085 .098 -.007 -.021
TRAV3E .722 .157 .078 .239 .083 .017 .042 .188
INDETIE .699 .091 .077 .189 .011 .171 .009 -.004
OWNR2E .696 .095 -.042 .136 .066 .109 -.029 .155
TRAV2E .685 .225 .054 .166 .097 .067 .023 .112
OWNR1E .680 .113 .000 .044 .082 .148 -.058 .089
EMRG3E .653 .269 .079 .117 -.067 .054 .018 .092
INDET3E .645 .182 .080 .176 .035 .074 .094 .217
CONV8E .272 .798 .086 .266 .011 .091 -.061 .131
CONV4E .240 .795 .038 .124 .087 .089 .025 .136
CONV7E .233 .779 .018 .279 -.007 .118 .034 .122
CONV3E .304 .768 .104 .150 .004 .044 -.037 .180
CONV6E .198 .767 -.026 .130 .082 .175 .036 .141
CONV5E .240 .723 -.021 .025 .141 .234 -.037 -.015
CONVIOE .288 .712 -.014 .347 .007 .014 -.003 .100
CONV9E .280 .684 -.070 .286 .137 .093 .092 .172
EOU2 .093 .029 .902 -.018 .068 -.034 -.058 .073
EOU6 .063 .048 .889 .014 .078 -.064 -.013 .036
EOU3 .110 .014 .884 .030 .058 -.049 -.023 -.010
EOU1 .106 -.023 .850 -.066 .046 -.013 .015 -.016
EOU5 .034 -.017 .836 .068 .088 .070 .163 .002
EOU4 .101 .036 .777 -.024 .116 -.124 -.093 .109
DIV8E .275 .387 -.047 .733 .039 .145 .094 .073
DIV4E .350 .317 .076 .638 .071 .240 .122 .116
DIV5E .306 .382 .058 .624 .056 .236 .038 .129
DIV7E .259 .502 -.021 .610 .119 .197 .106 .089
DIV6E .304 .331 -.072 .610 .110 .276 .108 .058
DIV3E .251 .341 .003 .590 .061 .317 -.028 .030
DIV9E .366 .496 .019 .569 .086 .129 .074 .072
COMM2E .076 .106 .047 .038 .828 .015 -.062 -.066
COMMIE .095 -.003 .009 -.078 .824 .217 .003 -.050
COMM5E .097 .074 .141 .031 .822 .028 .081 .037
COMM3E -.002 .044 .097 .146 .764 .137 .092 -.013
COMM4E .104 .092 .135 .097 .754 .001 -.033 .041
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 2
Table 5.1 continued
1 2 3 4 5 6 7 8
UMTGE .090 .081 -.045 .141 .082 .728 .011 .129
PROJE .204 .201 -.130 .248 .078 .676 -.060 .021
TELE .228 .179 -.083 .167 .146 .619 .071 .108
IPRE .299 .346 .066 .079 .107 .554 .014 -.042
BRIEFE .354 .076 -.075 .264 .088 .531 -.130 .187
MMEANS3 .040 .072 -.044 .061 .024 .027 .890 .104
MMEANS2 -.030 .037 .058 .019 .055 .018 .888 -.024
MMEANS1 .022 -.059 -.029 .108 -.011 -.069 .857 -.017
GLECTE .156 .125 .024 .092 -.013 .056 .031 .785
AICE .013 .210 .056 .162 -.020 .041 -.047 .744
TSEME .291 .102 -.059 .147 .019 .123 .049 .643
CORPE .181 .200 .212 -.189 -.056 .111 .047 .628
The second test of discriminant validity examines the average variance
extracted (AVE) for each construct. The AVE is a measure of the variance that a
construct captures from its items relative to the amount due to measurement error
(Chin, 1998b). AVE is calculated as follows: (ZA .;2) / ((ZAj2) + Z(1 - A i2)). In the
equation, A,, represents the factor loading for a construct’s item. The square root of
the AVE for the constructs should be larger than the correlations among constructs
(Chin, 1998b). This indicates that more variance is shared between the construct and
its items than with another construct comprised of a different set of items (Chin,
1998b). This was not the case for all constructs in the first wave data analysis.
The AVE for each construct is shown on the diagonal in Table 5.2. The
square root of the AVE for the perceived value of cross-departmental knowledge on
divergent thinking (.63) was less than the correlation between the perceived value of
cross-departmental knowledge on divergent thinking and convergent thinking (.74),
equal to the correlation between the perceived value of cross-departmental
knowledge on divergent thinking and personally initiated encounters (.63), and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
barely greater than the correlation between the perceived value of cross-departmental
knowledge on divergent thinking and the use of technology features that support
cross-departmental understanding (.61). The square root of the AVE for personally
initiated encounters (.63) was equal to the correlation between personally initiated
encounters and the perceived value of cross-departmental knowledge on divergent
thinking. The square root of the AVE for the perceived value of cross-departmental
knowledge on convergent thinking (.76) was just greater than the correlation
between the perceived value of cross-departmental knowledge on divergent thinking
and convergent thinking. In short, all of the variables did not show evidence of
discriminant validity, which means that support for the hypotheses would not be very
persuasive. However, first wave data analysis was continued to get a general sense
of the relationships among the variables under study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 5.2 First Wave Descriptive Statistics and Correlations
# of
Items a
h
s.d. 1 2 3 4 5 6 7 8
1 Perceived Value of Cross-departmental 7 .93 2.72 .88 .63
Knowledge on Divergent Thinking
2 Perceived Value o f Cross-departmental 8 .94 2.45 .86 .74** .75
Knowledge on Convergent Thinking
3 Perceived Openness o f Communication 5 .87 3.90 .71 .22**
19**
.80
4 Perceived Task Complexity 3 .87 4.06 .62 .15* .06 .05 .88
5 Corporate Initiated Encounters 4 .75 2.22 .73 .36**
41**
.03 .05 .70
6 Personally Initiated Encounters 5 .79 3.00 .82 .63** .50** .26** .01
30**
.63
7 Perceived Ease of Use of Technology 6 .93 3.26 .84 .04 .07 .19** .00 .12 -.06 .86
8 Use o f Technology Features that Support 13 .94 2.54 .84 .61** .57** .20** .04
37**
.51** .18** .71
Cross-departmental Understanding
N=248
**. Correlation is significant at the .01 level (2-tailed)
*. Correlation is significant at the .05 level (2-tailed)
The bolded numbers on the leading diagonal are the square root o f the variance shared between the constructs and their measures. Off diagonal elements are the
correlations among constructs. For discriminant validity, diagonal elements should be larger than off-diagonal elements.
oo
-I*-
85
5.1.2 First Wave Descriptive Statistics and Correlations
The descriptive statistics and correlations were examined for the six
independent variables and the two dependent variables (Table 5.2). Several points
about the correlations were noted. First, the two dependent variables were highly
correlated. This was expected. Recall from chapter 2, that innovation requires both
divergent thinking and convergent thinking. The positive correlation signaled that
survey respondents assessed the perceived value of cross-departmental knowledge on
innovation according to theory. In other words, the respondents believed that cross-
departmental knowledge helped both their divergent and convergent thinking or
helped neither. The respondents did not believe the value of cross-departmental
knowledge helped their divergent thinking, but not their convergent thinking (or visa
versa). Second, the two dependent variables were highly correlated with all the other
independent variables, except perceived ease of use of technology and perceived task
complexity. While these correlations may be evidence of true relationships, they
may also be the result of common method variance (section 5.1.5). Third, many of
the independent variables were also highly correlated with each other, with the
exception of perceived task complexity and perceived ease of use of technology.
This indicates that multicollinearity may exist among the set of predictors (Hair Jr.,
Anderson, Tatham, & Black, 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 6
5.1.3 Examination of First Wave Data Prior to Hierarchical Regression Analysis
Examination of the first wave data included identification of univariate
outliers. Whether the data met assumptions of linearity and normality was also
assessed. To check for univariate outliers, standardized scores (z-scores) were
created for the independent and dependent variables. Give that the sample size was
greater than 80, any z-score greater than or equal to 3.00 was flagged as a univariate
outlier. Four univariate outliers (case numbers 260, 392, 75, 790) were found for
perceived task complexity. Four respondents had low perceived task complexity
(2.00) compared to the mean (4.06). One univariate outlier (case number 287) was
found for corporate initiated encounters. The respondent used corporate initiated
encounters at a high rate (4.50) compared to the mean (2.22). Recognizing that a
certain number of observations naturally occur in these outer ranges of the
distribution, the cases were retained.
Prior to hierarchical regression analysis, the first wave data was also
examined for its adherence to assumptions of linearity and normality. To check
assumptions of linearity, scatter plots were examined (Hair Jr. et al., 1998). No
nonlinear relationships were apparent. To check assumptions of univariate
normality, results of graphical and statistical analyses were examined (Hair Jr. et al.,
1998). Evaluation of the histograms showed that the perceived value of cross-
departmental knowledge on convergent thinking was slightly positively skewed and
perceived openness of communication and perceived task complexity looked
negatively skewed. Normal probability plots showed that the perceived value of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
cross-departmental knowledge on divergent thinking and convergent thinking were
slightly kurtotic and perceived task complexity was negatively skewed. Using ±1.96
as the critical z-score value, skewness and kurtosis z-scores showed the perceived
value of cross-departmental knowledge on divergent thinking was slightly kurtotic
(z-score = -1.98), perceived task complexity was negatively skewed (z-score = -4.91)
and kurtotic (z-score = 4.75), and corporate initiated encounters was positively
skewed (z-score = 2.73). A one-sample Kolomogorov-Smimov test found all
variables violated assumptions of normality, with the exception of the use of
technology features that support cross-departmental understanding and the perceived
value of cross-departmental knowledge on convergent thinking.
In sum, five univariate outliers were identified but retained since observations
naturally occur in outer ranges of a distribution. Tests for linearity were conclusive;
no nonlinear relationships among variables were evident. However, results from
graphical and statistical analyses used to assess normality were equivocal. None of
the independent or dependent variables consistently exhibited departures in
normality, with the exception of perceived task complexity. Since hierarchical
regression analysis can be quite robust under violations of normality (Hair Jr. et al.,
1998), the perceived task complexity variable was not transformed. The soundness
of this decision was revisited when testing assumptions of linearity, normality,
heteroscedasticity, and independence of error terms for the final regression models.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 8
5.1.4 First Wave Hierarchical Regression Analysis
Two separate regressions were run to test hypotheses 1 and 2. For hypothesis
1, the perceived value of cross-departmental knowledge on divergent thinking was
the dependent variable, whereas the perceived value of cross-departmental
knowledge on convergent thinking was the dependent variable for hypothesis 2. In
both cases, the independent variable central to this dissertation study was the use of
technology features that support cross-departmental understanding. In addition, five
independent variables were controlled for: perceived openness of communication,
perceived task complexity, personally initiated encounters, corporate initiated
encounters, perceived ease of use of technology. Each hierarchical regression
proceeded in two steps. First, the base model (step 1) included the five independent
variables being controlled. Then the use of technology features that support cross-
departmental understanding was added to the base model (step 2). For the regression
on the perceived value of cross-departmental knowledge on divergent thinking and
the regression on the perceived value of cross-departmental knowledge on
convergent thinking, the change in R from step 1 to step 2 was examined to see if
there was a significant improvement in the model. In testing hypothesis 1 and 2, the
highest variance inflation factor was 1.52, which indicated that multicollinearity was
not a problem.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 9
Results in Table 5.3 indicate support for hypothesis 1. The use of technology
features that support cross-departmental understanding was positively related to the
perceived value of cross-departmental knowledge on divergent thinking after
controlling for the other five variables. The F statistic (p < .001) showed the final
model to be significant. The change in R2 was also significant (p < .001). The
change indicated that adding the use of technology features that support cross-
departmental understanding to the regression equation went toward explaining a
significant amount of additional variance (8.8%) in the perceived value of cross-
departmental knowledge on divergent thinking. Further examination of the beta
coefficients also showed that personally initiated encounters (p < .001), corporate
initiated encounters (p < .05), and perceived task complexity (p < .01) were also
positively related to the perceived value of cross-departmental knowledge on
divergent thinking. In contrast, perceived openness of communication and perceived
ease of use of technology were not significantly related to the perceived value of
cross-departmental knowledge on divergent thinking. Post regression examination
of the residual plot showed no patterns, which confirmed assumptions of linearity
and homoscedasticity. In addition, the normal probability plot of residuals and
histogram plot of residuals showed no systematic or substantial departures from the
assumption of normality. Lastly, the Durbin-Watson statistic (2.02) confirmed that
the error terms were independent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 0
Table 5.3 First Wave Hierarchical Regression Results
For Perceived Value of
Cross-departmental
Knowledge on
Divergent Thinking
For Perceived Value of
Cross-departmental
Knowledge on
Convergent Thinking
Step 1 Step 2 Step 1 Step 2
Control Variables |3:
Intercept .000 .000 .000 .000
Perceived Openness of Communication .059 .045 .063 .049
Perceived Task Complexity .131** .123** .034 .025
Corporate Initiated Encounters .175** .099*
278***
.201***
Personally Initiated Encounters .558*** .395***
403*** 237***
Perceived Ease of Use of Technology .041 -.021 .047 -.016
Focal Independent Variable (1 :
Use of Technology Features that .366***
372***
Support Cross-departmental Understanding
Model Statistics:
N 248 248 248 248
R2 44.6% 53.4% 33.3% 42.3%
Adjusted R2 43.4% 52.2% 31.9% 40.9%
Model F 38.92*** 45 97*** 24.J2***
29.49***
AR 2 8.8% 9.1%
F for A R2
45 47*** 37 92***
p < .15, A p < .10, *p < .05, **p < .01, ***p < .001
Results in Table 5.3 indicate support for hypothesis 2. The use of technology
features that support cross-departmental understanding was positively associated
with the perceived value of cross-departmental knowledge on convergent thinking,
after controlling for the other five variables. The F statistic for the final model was
significant (p < .001). There was also a significant change in the R2 (9.1%). Adding
the use of technology features that support cross-departmental understanding to the
regression equation explained a significant amount of additional variance in the
perceived value of cross-departmental knowledge on convergent thinking (p < .001).
Other significant variables that were positively related to the perceived value of
cross-departmental knowledge on convergent thinking were personally initiated
encounters (p < .001) and corporate initiated encounters (p < .001). While perceived
task complexity was significantly related to the perceived value of cross-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
departmental knowledge on divergent thinking, the same cannot be said of its
relationship to the perceived value of cross-departmental knowledge on convergent
thinking. In addition, perceived openness of communication and perceived ease of
use of technology were not significantly related to the perceived value of cross-
departmental knowledge on convergent thinking. The post regression analysis of
residual plots indicated that the assumptions of linearity, homoscedasticity, and
normality were met. The Durbin-Watson statistic (1.92) confirmed that the error
terms were independent.
5.1.5 Two Major Limitations of First Wave Data Analysis
While the results of the first wave data analysis supported hypotheses 1 and
2, the collection and analysis of first wave data presented two major limitations -
common method variance and discriminant validity. Common method variance was
introduced when the data for the independent and dependent variables was collected
from the same person at the same time (Podsakoff et al., 2003). This type of survey
design can increase the likelihood of a strong relationship between the independent
and dependent variables, as seen in Table 5.2. When respondents are filling out
surveys that measure independent and dependent variables simultaneously, several
things can occur to bias the results. For example, respondents’ answers to earlier
questions may remain in their short-term memory and provide retrieval cues when
answering later questions (Podsakoff et al., 2003). In addition, respondents’ answers
to previous questions may provide a context that allows them to create and use
implicit theories during survey completion (Podsakoff et al., 2003). In both cases, an
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 2
association may be created between independent and dependent variables that may
not exist if the two sets of variables were collected at separate times. In short, the
significance of the predictors in hypotheses 1 and 2 (Table 5.3) may be due in part to
the common measurement context provided during survey administration.
In the case of this dissertation study, the common method variance also may
have negatively impacted the tests of discriminant validity among the constructs.
The presence of common method variance, makes it difficult to disentangle whether
the issues related to discriminant validity are due to poor survey items or the method
effects. Given the results from the preliminary data collection activities, survey
pilots, and factor analyses, it was suspected that the lack of discriminant validity was
due to method effects not poor survey items. A second wave survey collected
dependent variable data only to address these limitations. With the time separation
between the independent and dependent variables, respondents were not be able to
fill in memory gaps, create and use implicit theories, or rely on retrieval cues
(Podsakoff et al., 2003). It was also expected that the second wave data collection of
the dependent variables, in the absence of priming effects from the independent
variables, would enable respondents to make better distinctions when it came to
deciding whether cross-departmental knowledge helped their divergent thinking and
convergent thinking. Although theory suggests that the correlation between these
two constructs will remain high, the items for the perceived value of cross-
departmental knowledge on divergent thinking and convergent items may load
higher on their own construct, which would boost the AYE.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 3
5.2 Second Wave Data Analysis
Matching respondents’ independent variables collected from the first wave to
their dependent variables collected during the second wave created a second wave
data set. The separate collection of independent and dependent variables addressed
the issue of common method variance. As expected, the second wave data analysis
showed discriminant validity among the constructs (section 5.2.1). The descriptive
statistics and correlations for the second wave data analysis are presented (section
5.2.2), followed by results from a preliminary examination of the data (section 5.2.3)
and hypothesis testing (section 5.2.4), including identification of outliers (section
5.2.5), and a retest of hypotheses with outliers removed (section 5.2.6).
5.2.1 Second Wave Tests for Discriminant Validity
The first test of discriminant validity subjected all items for all constructs to a
principal components factor analysis with varimax rotation. The final results are
shown in Table 5.4. Three items (EMRG1E, TELE, CORPE) were eliminated due to
cross loadings.
Table 5.4 Second Wave Factor Solution with Varimax Rotation
1 2 3 4 5 6 7 8
SCONV7E .872 .177 .213 .169 .069 .081 .048 .059
SCONV3E .851 .217 .247 .114 .025 .005 -.054 .045
SCONV4E .847 .187 .244 .125 .051 .080 .037 .051
SCONVIOE .823 .142 .323 .134 -.063 -.059 .035 .086
SCONV8E .815 .176 .261 .097 .028 .025 .109 .049
SCONV5E .805 .120 .337 .000 .035 .094 -.035 .068
SCONV6E .792 .243 .325 .113 .059 .024 .073 .009
SCONV9E .779 .187 .359 .086 .005 .026 .062 .076
SCONV2E .750 .087 .452 -.009 .012 .030 -.031 .008
SCONV1E .739 .092 .294 .131 .076 .094 .010 .122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 4
Table 5.4 continued
1 2 3 4 5 6 7 8
INDET2E .052 .796 .045 .077 .086 -.066 .192 -.036
MULT2E .238 .755 .141 .062 .008 .091 .086 -.126
EMRG3E .069 .733 .084 -.007 -.053 -.077 -.020 .012
TRAV1E .147 .729 .235 .047 .105 .052 -.081 .088
MULT3E .180 .727 -.017 -.070 .064 -.064 .198 -.157
OWNR3E .054 .720 .090 .051 -.112 -.167 .081 .183
MULT1E .272 .719 .145 .046 .054 .049 .056 -.003
TRAV2E .124 .717 .105 .042 .002 .057 .197 .070
INDETIE .112 .715 .149 .034 .095 -.050 .1 1 1 -.006
TRAV3E .152 .715 .259 .121 .115 -.020 .032 .158
OWNR2E .046 .700 .291 .105 .067j -.014 -.005 .214
OWNR1E -.016 .675 .030 .091 -.032 -.008 .048 .207
INDET3E .216 .672 .192 .062 -.018 .019 .086 .137
EMRG2E .106 .668 .018 .044 .067 .063 .125 -.073
SDIV8E .394 .180 .797 -.058 .019 -.033 .047 -.073
SDIV4E .336 .245 .789 .005 .048 .089 .030 .071
SDIV5E .387 .143 .786 -.004 .004 .048 .100 .002
SDIV9E .386 .195 .774 .038 .009 .008 .137 .049
SDIV3E .372 .155 .767 .037 -.012 .071 .092 .037
SDIV6E .369 .237 .765 -.009 .066 -.027 .059 -.091
SDIV2E .249 .145 .762 .067 .120 .015 .066 .117
SDIV7E .414 .195 .751 -.029 .083 .005 .175 -.018
SDIV1E .414 .192 .704 .035 .042 .103 .120 .023
EOU2 .091 .101 .054 .879 .068 -.082 -.004 .031
EOU3 .122 .078 .1 1 1 .860 .081 -.065 .029 -.110
EOU6 .057 .067 -.068 .841 .163 -.008 -.076 .054
EOU5 .097 .050 .052 .833 .131 .157 .117 .012
EOU1 .122 .113 .0 1 1 .774 .141 .079 .093 -.087
EOU4 .148 .061 -.097 .711 .174 -.104 -.209 -.035
COMMIE .045 .118 .083 .005 .824 -.050 .185 -.062
COMM2E -.040 .028 .055 .067 .797 -.073 -.140 -.052
COMM5E .074 .019 .049 .220 .792 .141 .058 -.040
COMM4E .107 .065 .018 .279 .694 -.026 -.007 .091
COMM3E .005 .014 .030 .231 .686 .079 .269 .034
MMEANS3 .087 -.028 .090 .008 -.074 .882 .002 .055
MMEANS2 .131 -.036 .048 .018 .038 .867 .066 .013
MMEANS1 .021 -.046 .007 -.043 .068 .788 .054 -.118
PROJE .015 .238 .174 -.071 .080 .002 .767 .061
IPRE .207 .276 .037 .060 .089 .067 .660 -.067
UMTGE -.040 .163 .216 -.037 .160 .092 .551 .194
BRIEFE -.023 .383 .180 .040 -.031 -.003 .532 .339
GLECTE .085 .095 .104 .007 -.016 .047 .101 .739
TSEME .069 .259 -.032 -.087 -.038 .051 -.041 .739
AICE .214 -.039 -.045 -.059 .019 -.209 .181 .725
Telephone (TELE) was one of the items intended to measure the use of
personally initiated encounters. However, the item’s factor loading for its construct
was low (.305), and the item cross-loaded with items measuring the use of
technology features that support cross-departmental understanding, the perceived
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 5
value of cross-departmental knowledge on divergent thinking, perceived openness of
communication, and corporate initiated encounters. It may be that telephone use is
too ubiquitous to be used as a measure of any one phenomenon. The fact that the
telephone is used under so many circumstances may be why it was an indicator on
more than one construct.
One item from the use of technology features that support cross-departmental
understanding (EMRG1E) was also deleted because it cross-loaded with items from
the personally initiated encounters construct. In retrospect, the wording of the item
may have caused this to occur. The item is stated as follows, “find changes made to
the knowledge as ideas evolve over time (e.g. access current and historical program
data).” It may be that people use personally initiated encounters to find out what
changes have been made to knowledge (i.e. telephone a colleague from a prior
project) as well as technology (e.g. email a colleague).
The use of corporate events (CORPE) also had a low factor loading (.488) on
its intended construct - corporate initiated encounters. A poor choice of words may
have been the issue. Even though an example was provided, the term (i.e. corporate
events) may have been too ambiguous in comparison to other items measuring the
use of corporate initiated encounters (e.g. technical seminars, guest lectures,
Company A institute courses). As shown on the diagonal of Table 5.5, results of the
second test of discriminant validity showed that the square root of the AVE for the
constructs was higher than the correlations among constructs. Problems with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 6
discriminant validity identified during the first wave data analysis were not an issue
for the second wave.
5.2.2 Second Wave Descriptive Statistics and Correlations
Continuing with Table 5.5, two observations with respect to the correlations
among variables follow. First, as expected, the perceived value of cross-departmental
knowledge on divergent thinking and the perceived value of cross-departmental
knowledge on convergent thinking were still highly correlated, supporting theory
that innovation requires both. Second, the correlations between the dependent
variables from the second wave data collection and the independent variables from
the first wave data collection were still significant in most cases. However, the
correlations in the second wave were less than those of the first wave.
Table 5.5 also shows descriptive statistics, including the change in
reliabilities after the elimination of items in the personally initiated encounters (.76
to .72) and corporate initiated encounters constructs (.72 to .68). Although the
change in reliability for the corporate initiated encounters construct fell below .70, it
was thought best to eliminate an ambiguously worded item rather than maintain
reliability above .70. Furthermore, at a = .68, the drop in the reliability was neither
overly large nor far below the .70 threshold. There was no change in reliability in
the use of technology features that support cross-departmental understanding.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Tab] e 5.5 Second Wave Descriptive Statistics and Corre ations
# o f
Items a
F
s.d. 1 2 3 4 5 6 7 8
l Perceived Value of Cross-departmental 9 .96 2.82 .91 .77
Knowledge on Divergent Thinking
2 Perceived Value of Cross-departmental 10 .97 2.27 .88
y2**
.81
Knowledge on Convergent Thinking
3 Perceived Openness o f Communication 5 .84 3.93 .64 .14 .13 .76
4 Perceived Task Complexity 3 .83 4.13 .54 .12 .14 .04 .84
5 Corporate Initiated Encounters 3 .68 2.34 .77 .14 .21* -.02 -.06 .73
6 Personally Initiated Encounters 4 .72 2.84 .82
,23**
.21* .10
27**
.64
7 Perceived Ease of Use o f Technology 6 .91 3.36 .77 .10
24**
.35** .00 -.04 .03 .82
8 Use of Technology Features that Support 14 .94 2.52 .80
43**
40** .14 -.03 .21*
48**
.18* .72
Cross-departmental Understanding
N=131
**. Correlation is significant at the .01 level (2-tailed)
*. Correlation is significant at the .05 level (2-tailed)
The bolded numbers on the leading diagonal are the square root of the variance shared between the constructs and their measures. Off diagonal elements are the
correlations among constructs. For discriminant validity, diagonal elements should be larger than off-diagonal elements.
vo
- J
9 8
5.2.3 Examination of Second Wave Data Prior to Hierarchical Regression
To check for univariate outliers, standardized scores (z-scores) were created
for the independent and dependent variables. Given that the sample size was greater
than 80, any z-score greater than or equal to 3.00 was flagged as a univariate outlier.
One univariate outlier was found for perceived ease of use of technology (case
number 729), perceived task complexity (case number 63), the perceived value of
cross-departmental knowledge on convergent thinking (case number 528), and
corporate initiated encounters (case number 566). Recognizing that a certain number
of observations naturally occur in these outer ranges of the distribution, the cases
were retained.
Prior to hierarchical regression analysis, the second wave data was examined
for its adherence to assumptions of linearity and normality. Scatter plots showed no
indication of nonlinear relationships among variables. Histograms showed the
perceived value of cross-departmental knowledge on convergent thinking and
corporate initiated encounters were positively skewed, whereas perceived ease of use
of technology, perceived openness of communication, and perceived task complexity
were negatively skewed. Normal probability plots did not show any major
departures from normality. Using ±1.96 as the critical z-score value, skewness and
kurtosis z-scores showed the perceived value of cross-departmental knowledge on
convergent thinking (z-score=2.15) was positively skewed. A one-sample
Kolomogorov-Smimov test found perceived task complexity and corporate initiated
encounters had significant departures from normality. In brief, evidence supported
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 9
assumptions of linearity, however evidence supporting assumptions of normality,
was inconclusive. Additional tests of linearity, normality, heteroscedasticity, and
independence of error terms done on the final regression models found that there
were no systematic or substantial departures from these assumptions.
5.2.4 Second Wave Hierarchical Regression Analysis
Similar to the first wave data analysis, two separate regressions were run to
test hypotheses 1 and 2. The perceived value of cross-departmental knowledge on
divergent thinking was the dependent variable for hypothesis 1 and the perceived
value of cross-departmental knowledge on convergent thinking was the dependent
variable for hypothesis 2. Step 1 included the five independent variables being
controlled - perceived openness of communication, perceived task complexity,
personally initiated encounters, corporate initiated encounters, and perceived ease of
use of technology. Step 2 added the independent variable central to this dissertation
study - the use of technology features that support cross-departmental understanding.
For both regressions, the change in R2 from step 1 to step 2 was examined to see if
there was a significant improvement in the model. The highest variance inflation
factor for the regression models used to test hypothesis 1 and hypothesis 2 were both
1.43, which indicated that multicollinearity was not a problem.
Results in Table 5.6 indicate support for hypothesis 1. The use of technology
features that support cross-departmental understanding was positively related to the
perceived value of cross-departmental knowledge on divergent thinking after
controlling for the other five variables. The F statistic (p < .001) showed the final
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 0
model to be significant. The 8.3% change in R2 was also significant (p < .001).
Adding the use of technology features that support cross-departmental understanding
to the regression equation explained a significant amount of additional variance in
the perceived value of cross-departmental knowledge on divergent thinking.
Although none of the other independent variables were shown to be significantly
related to the perceived value of cross-departmental knowledge on divergent
thinking, personally initiated encounters was positively related to the perceived value
of cross-departmental knowledge on divergent thinking (p < .10). Post regression
examination of the residuals plots, including the normal probability plot and
histogram confirmed assumptions of linearity, homoscedasticity, and normality. The
Durbin-Watson statistic (2.16) confirmed that the error terms were independent.
Table 5.6 Second Wave Hierarchical Regression Results
For Perceived Value of
Cross-departmental
Knowledge on
Divergent Thinking
For Perceived Value of
Cross-departmental
Knowledge on
Convergent Thinking
Step 1 Step 2 Step 1 Step 2
Control Variables 3:
Intercept .000 .000 .000 .000
Perceived Openness of Communication .046 .049 .014 .018
Perceived Task Complexity .089 .111 .137* ,159A
Corporate Initiated Encounters .053 .022 .190* .159A
Personally Initiated Encounters 222*** ,179A .153A .001
Perceived Ease of Use of Technology .073 .015 .238** .180*
Focal Independent Variable 3:
Use of Technology Features that 227*** .334***
Support Cross-departmental Understanding
Model Statistics:
N 131 131 131 131
R2 15.3% 23.7% 15.4% 23.6%
Adjusted R2 11.9% 20.0% 12.0% 19.9%
Model F 4.53** 6.40*** 4.56** 6.38***
A R 2 8.3% 8.2%
F for A R2 13.51*** 13.28***
*p < .15, A p < .10, *p < .05, **p < .01, ***p < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
Results in Table 5.6 indicate support for hypothesis 2. The use of technology
features that support cross-departmental understanding was positively associated
with the perceived value of cross-departmental knowledge on convergent thinking,
after controlling for the other five variables. The F statistic for the final model was
significant (p < .001) and there was a significant change in the R2 (8.2%). Perceived
ease of use of technology was also positively related to the perceived value of cross-
departmental knowledge on convergent thinking (p < .05). Perceived task
complexity and corporate initiated encounters were positively associated with the
perceived value of cross-departmental knowledge on convergent thinking (p < .10).
A post regression analysis of residual plots, including the normal probability plot and
histogram indicated assumptions of linearity, homoscedasticity, and normality were
met. Error terms were found to be independent given the Durbin-Watson statistic
(2.11).
5.2.5 Identification of Influential Second Wave Data Points
Second wave data analysis included several diagnostic tests to identify
outliers, leverage points, and influential observations for each regression equation
(Hair Jr. et al., 1998). Table 5.7 lists the threshold value specification, calculated
threshold value, and the top five observations for each diagnostic measure.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 2
Table 5.7 Second Wave Influential Observations
Top 5 Influential Points for:
Diagnostic
Measure
Threshold
Value
Specification
Calculated
Threshold
Value
Perceived Value of
Cross-departmental
Knowledge on
Divergent Thinking
Perceived Value of
Cross-departmental
Knowledge on
Convergent Thinking
Residuals: Critical
t-value at
specified
confidence
level
±1.96
Standardarized 154,367,659,528,793 192,528,787,651
Studentized 154,367,528,659,793 192,528,787,651
Studentized Deleted 154,367,528,659,793 192,528,787,651
Leverage Values
Centered leverage
value
2(k+l)/n .11 658,566,63,584,775 658,566,63,584,775
Mahalanobis distance
value
Top 5
observations
658,566,63,584,775 658,566,63,584,775
Single Case Measures:
Cook’s distance 4/(n-k-l) .03 528,775,659,793,270 528,775,651,787,584
COVRATIO 1 ± 3(k+l)/n Upper: 1.16
Lower: .84
154,566,63,367,729 192,566,729,63,740
SDFFIT 2 * [sqroot
of (k+l)/(n-
k-1)
.48 528,775,659,793,270 528,775,651,787,584
SDFBETA ±2/(sqroot of
n)
±.17
Intercept 154,367,528,659,793 528,192,651,787
Perceived Ease of Use
of Technology
775,393,781,765,270 775,848,170,651,528
Perceived Task
Complexity
528,775,270,801,808 528,775,612,63,784
Perceived Openness of
Communication
659,367,528,352,270 528,775,352,703,289
Personally Initiated
Encounters
528,154,470,651,351 528,651,787,351,170
Corporate Initiated
Encounters
793,659,367,787,393 793,787,177,584,843
Use of Technology
Features that Support
Cross-departmental
Understanding
528,351,154,781,270 528,584,351,848,220
n = 131 (sample size), k = 7 (number of independent variables)
First, residuals were examined to identify the top five outliers on the
dependent variables. Next, centered leverage values and Mahalanobis distance
values were used to identify the top five leverage points for each regression equation.
Leverage points are observations that are considerably different than all other
observations on one or more independent variables (Hair Jr. et al., 1998). Lastly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 3
Cook’s distance, the covariance ratio, standardized DFFit, standardized DFBetas
estimate the influence a single case has on the regression results (Hair Jr. et al.,
1998). Cases that are consistently identified as influential points across the
diagnostic tests are the most likely to impact the regression model (Hair Jr. et al.,
1998). Since two separate regression equations were tested for this dissertation
study, diagnostics were calculated for each. Two cases (528 and 775) were
consistently identified in both sets of diagnostic measures as likely candidates for
deletion.
5.2.6 Revisiting Hierarchical Regression Analysis with Influential Points Removed
The two cases represent extremes in comparison to the means in Table 5.5,
For example, the respondent for case number 775 used technology features that
support cross-departmental understanding (1.29) and personally initiated encounters
(1.50) less than average, perceived technology to be easier to use than average
(4.67), perceived tasks to be more complex than average (5.00), perceived
communication to be less open than average (3.00), and perceived the value of cross-
departmental knowledge on divergent thinking (1.00) and convergent thinking (1.00)
to be less than average. In contrast, the respondent for case 528, used technology
features that support cross-departmental understanding more than average (3.36),
used personally initiated encounters less than average (1.50), perceived tasks to be
more complex than average (5.00), perceived communication to be more open than
average (5.00), and perceived cross-departmental knowledge to help divergent
thinking (4.89) and convergent thinking (5.00) more than average.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 4
To consider how influential these points might be, hierarchical regression
analysis was performed without them to see whether any change occurred.
Hypothesis 1 was still supported (Table 5.8). The F statistic for the final model
remained significant (p < .001). Even though the change in R2 dropped to 6.2%, the
positive association between the use of technology features that support cross-
departmental understanding and the perceived value of cross-departmental
knowledge on divergent thinking remained significant (p < .01). In this model, the
personally initiated encounters construct was also found to positively relate to the
perceived value of cross-departmental knowledge on divergent thinking (p < .05).
This is a change from the prior model, which showed the relationship to be
significant at p < .10. Hypothesis 2 was also still supported (Table 5.8). The F
statistic for the final model remained significant (p < .001) and the 5.9% change in
R2 was significant (p < .01). The positive relationship between the use of technology
features that support cross-departmental understanding and the perceived value of
cross-departmental knowledge on convergent thinking remained significant (p < .01).
The positive association between perceived ease of use of technology and the
perceived value of cross-departmental knowledge on convergent thinking also
remained significant (p < .01). Corporate initiated encounters and perceived task
complexity also remained positively related to the perceived value of cross-
departmental knowledge on convergent thinking (p < .10).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 5
Table 5.8 Second Wave Hierarchical Regression Results Without Influential Points
For Perceived Value of
Cross-departmental
Knowledge on
Divergent Thinking
For Perceived Value of
Cross-departmental
Knowledge on
Convergent Thinking
Step 1 Step 2 Step 1 Step 2
Control Variables P:
Intercept -.007 -.005 -.008 -.006
Perceived Openness of Communication -.018 -.003 -.055 -.041
Perceived Task Complexity .077 .098 .124" .145A
Corporate Initiated Encounters .039 .015 .174* .151A
Personally Initiated Encounters .226* .185* .054
Perceived Ease of Use of Technology .088 .036 .254** .204*
Focal Independent Variable P:
Use of Technology Features that .285** .276**
Support Cross-departmental Understanding
Model Statistics:
N 129 129 129 129
R2 16.7% 22.8% 16.5% 22.4%
Adjusted R2 13.3% 19.0% 13.1% 18.6%
Model F
4 9i***
6.01*** 4.85*** 5.87***
AR2 6.2% 5.9%
F for A R2 9.75** 9.29**
*p < .15, A p < .10, *p < .05, **p < .01, ***p < .001
In sum, two influential points common to both regression equations were
identified. Performing hierarchical regression with these two points removed found
that hypothesis 1 was still supported. In addition, for the regression on the perceived
value of cross-departmental knowledge on divergent thinking, the personally
initiated encounters variable was significant (p < .05). Hypothesis 2 was also still
supported, no changes were found for the regression on the perceived value of cross-
departmental knowledge on convergent thinking. Although there was a slight
change in the results of the final regression equation for the perceived value of cross-
departmental knowledge on divergent thinking, the change was not enough to
warrant the permanent removal of these two cases from the data set. The final
results will include them.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 6
5.3 Hierarchical Regression Analysis for the Exploratory Question
Recall from chapter 4, that an exploratory question was posed. More
specifically, five interactive effects between the use of technology features that
support cross-departmental understanding, and a) perceived openness of
communication, b) perceived task complexity, c) personally initiated encounters, d)
corporate initiated encounters, e) perceived ease of use of technology, were tested.
Using the full second wave data set, these interactions were tested in a regression on
the perceived value of cross-departmental knowledge on divergent thinking and a
regression on the perceived value of cross-departmental knowledge on convergent
thinking.
Following Aiken and West (1991), the five interactions were created by
multiplying the z-scores of the variables involved. For both regressions step 1 and
step 2 remained the same as the prior hierarchical regression analyses. Step 1
included the z-scores for: perceived openness of communication, perceived task
complexity, personally initiated encounters, corporate initiated encounters, perceived
ease of use of technology. Step 2 included the z-score for the use of technology
features that support cross-departmental understanding. Step 3 was added to include
the five interactions.
For both regressions the change in R2 was examined to see if there was
significant improvement in the model. For the regression on the perceived value of
cross-departmental knowledge on divergent thinking there was no significant change
in R2; neither the F statistic nor the beta coefficients were significant (Table 5.9).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 7
The same held true for the regression on the perceived value of cross-departmental
knowledge on convergent thinking. The interaction closest to significance was
found in the regression on the perceived value of cross-departmental knowledge on
convergent thinking (Table 5.10). The interaction between the use of technology
features that support cross-departmental understanding and the perceived ease of use
of technology was positively related to the perceived value of cross-departmental
knowledge on convergent thinking (p < .15).
Table 5.9 Second Wave Exploratory Hierarchical Regression for Perceived Value of
Cross-departmental Knowledge on Divergent Thinking
For Perceived Value of Cross-
departmental Knowledge on
Divergent Thinking
Step 1 Step 2 Step 3
Control Variables 3:
Intercept .000 .000 .027
Perceived Openness of Communication .046 .049 .006
Perceived Task Complexity .089 .1 1 1 .118
Corporate Initiated Encounters .053 .022 -.004
Personally Initiated Encounters
332***
,179A ,187A
Perceived Ease of Use of Technology .073 .015 .056
Focal Independent Variable 3:
Use of Technology Features that Support Cross-
departmental Understanding
337***
.306**
Interaction Variables 3:
Use of Technology Features that Support Cross-
departmental Understanding * Personally Initiated Encounters
-.092
Use of Technology Features that Support Cross-
departmental Understanding* Corporate Initiated Encounters
.012
Use of Technology Features that Support Cross-
departmental Understanding * Perceived Ease of Use of
Technology
.063
Use of Technology Features that Support Cross-
departmental Understanding * Perceived Openness of
Communication
.031
Use of Technology Features that Support Cross-
departmental Understanding * Perceived Task Complexity
.070
Model Statistics:
N 131 131 131
R2 15.3% 23.7% 25.2%
Adjusted R2 11.9% 20.0% 18.2%
Model F 4.53** 6.40*** 3.68***
A R 2 8.3% 1.5%
F for A R2 13.51*** .482
'p < .15, A p < .10, *p < .05, **p < .01, ***p < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 8
Table 5.10 Second Wave Exploratory Hierarchical Regression for Perceived Value
of Cross-departmental Knowledge on Convergent Thinking
For Perceived Value of Cross-
departmental Knowledge on
Convergent Thinking
Step 1 Step 2 Step 3
Control Variables (3 :
Intercept .000 .000 -.016
Perceived Openness o f Communication .014 .018 -.032
Perceived Task Complexity .137# .159A .168*
Corporate Initiated Encounters .190* .159A .115
Personally Initiated Encounters .153A .001 .005
Perceived Ease of Use of Technology .238** .180* .218*
Focal Independent Variable P:
Use o f Technology Features that Support Cross-
departmental Understanding
.334*** .289**
Interaction Variables P:
Use of Technology Features that Support Cross-
departmental Understanding * Personally Initiated Encounters
-.034
Use of Technology Features that Support Cross-
departmental Understanding * Corporate Initiated Encounters
.049
Use of Technology Features that Support Cross-
departmental Understanding * Perceived Ease of Use of
Technology
.142*
Use o f Technology Features that Support Cross-
departmental Understanding * Perceived Openness of
Communication
-.005
Use of Technology Features that Support Cross-
departmental Understanding * Perceived Task Complexity
.090
Model Statistics:
N 131 131 131
IF 15.4% 23.6% 26.1%
Adjusted R2 12.0% 19.9% 19.2%
Model F 4.56** 6.38*** 3 32***
AR2 8.2% 2.5%
F for A R2 13.28*** .80
*p < .15, A p < .10, *p < .05, **p < .01, ***p < .001
5.4 Summary
Two waves of data were analyzed. Although the analysis of the first wave
data supported hypotheses 1 and 2, two major limitations compromised the findings.
First, common method variance was introduced when the independent and dependent
variables were collected from the same person at the same time. Second, tests
showed discriminant validity was not supported for the perceived value of cross-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 9
departmental knowledge on divergent thinking or personally initiated encounters. A
second wave of data collection, for the dependent variables only, addressed the issue
of common method variance. In addition, analysis of the second wave data showed
that discriminant validity was not a problem. Hierarchical regression analysis of the
second wave data showed support for hypothesis 1 and 2 in the presence and absence
of two influential points. The influential points were retained. Additional tests to
explore the interactive effects between the use of technology features that support
cross-departmental understanding and other factors in the environment were
conducted. None of the interactive effects were significant. The next chapter
discusses the theoretical and practical implications of this dissertation study as well
as ideas for future research.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
CHAPTER 6: DISCUSSION
Notwithstanding the limitations of this dissertation study (section 6.1), this
chapter highlights several contributions to research (section 6.2), ideas for future
study (section 6.3) as well as contributions to practice (section 6.3). The chapter
ends with a summary of the research findings and conclusions (section 6.4).
6.1 Limitations
As with any research, this dissertation study is not without limitations. Those
that are most notable include the data collection procedures, which took place at one
site (section 6.1.1), and employed a two-wave survey methodology that reduced the
response rate (section 6.1.2). In addition, two of the major survey instruments were
newly developed for this dissertation study (section 6.1.3).
6.1.1 Data Collection Procedures Took Place at One Site
All of the data collection activities for this dissertation study took place at
one site. Thus, the generalizability of the research findings may be limited. For
example, most Company A employees interviewed and surveyed for this dissertation
study have advanced degrees in the engineering and science disciplines. In addition,
the findings indicated that most employees agreed that their work was complex, the
communication across departments was open, and KM technologies were easy to
use. Under these circumstances, technology features that support cross-departmental
understanding may be better received. Despite the limitations due to data collection,
the findings are still instructive. This is especially true considering the continued
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
challenge organizations face getting employees to seek, share, or otherwise transfer
each other’s knowledge across organizational boundaries (e.g. Carlile, 2002;
Dougherty, 1992; Markus et al., 2002) and the recent calls for the IS research
community to examine alternative designs, uses, and successes of technology (e.g.
Alavi & Leidner, 2001; Benbasat & Zmud, 2003; Markus, 2001; Orlikowski &
Iacono, 2001).
6.1.2 Two-wave Survey Methodology Reduced the Response Rate
The use of a two-wave survey methodology reduced the response rate to 15%
of the sample population. The small percentage calls the representativeness of the
data sets used in the first and second wave analyses into question. However, chi-
square tests showed that non-response bias was not a problem in the first or second
wave data sets. In addition, the sample sizes for the first (N=248) and second
(N=131) wave data sets were large enough to perform hierarchical regression
analysis and test for medium effect sizes. Given no evidence of non-response bias
and a large enough sample size to adequately test the hypotheses, the need for a high
response rate becomes a moot point. Moreover, the two-wave survey methodology
addressed the issue of common method variance bias by collecting the independent
and dependent variables at different points in time.
6.1.3 Two Survey Instruments Were Newly Developed
The third limitation is the use of two newly developed survey instruments.
No prior survey instrument existed to accurately capture the way employees accessed
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
cross-departmental knowledge. Thus, survey instruments for the use of technology
features that support cross-departmental understanding and people-to-people
interaction constructs were specifically developed for this dissertation study. While
this approach may come at the expense of the replicability of this dissertation
research, these new survey instruments were theoretically derived and refined during
survey pilots. In addition, the results of statistical analysis showed the measures to
be valid and reliable. Thus, using these instruments for future research is possible.
6.2 Contributions to Theory
Despite this dissertation study’s limitations, the findings from this research
provide several contributions to theory. First, the findings showed that KM
technologies could be used to help bridge the barriers to cross-departmental
knowledge access. KM technologies have been touted as a means to widely
disseminate diverse knowledge across organizational boundaries (e.g. Alavi &
Leidner, 2001; Huber, 1990). However, much of the research to date has shown that
using KM technologies under such circumstances is problematic (Alavi & Leidner,
2001; Markus, 2001). For example, Goodman and Darr (1998) found KM
technology to be of little use to individuals solving heterogeneous problems. It was
inferred that the KM technology was not able to adequately support the
heterogeneity of knowledge required during problem solving. Moreover, prior
research has argued that the codification and storage of knowledge via KM
technologies offers poor support to individuals needing to meaningfully interpret and
adapt specialized knowledge (Hansen et al., 1999; Nonaka & Takeuchi, 1995),
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 3
especially when the knowledge comes from an area of expertise that is different than
their own. Dixon (2000) has suggested that the benefits associated with KM
technology are limited to the exchange of knowledge among people with shared or
similar interests. In contrast, this dissertation study suggests that technology features
can bridge the barriers to cross-departmental knowledge access by providing access
to knowledge that is: a) associated with the expert source, b) interpretable, and c)
adaptable. Specifically, findings from this dissertation study suggest that cross-
departmental knowledge is more highly valued when technology features support
knowledge that: allows individuals to make their biases known, provides
contradictory alternative solutions, provides links to additional contextual details,
and is open to unpredictable change.
The findings from this dissertation study also contribute to the boundary
object literature. Boundary objects are entities that “inhabit several intersecting
social worlds and satisfy the information requirements of each of them” (Star &
Griesemer, 1989 p. 393). Several types of boundary objects have been described for
their role in supporting access to the diversity of knowledge that exists among
individuals from different thought worlds (Carlile, 2002; Henderson, 1991; Star &
Griesemer, 1989). For example, Carlile notes that effective boundary objects are
those that help people represent their own knowledge, learn about other people’s
knowledge, and transform their own and other’s knowledge. Similarly, prior
boundary object research has suggested several types of boundary objects (e.g.
repositories, standard forms and methods, objects, models, maps) that can help (e.g.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 4
Carlile 2002; Star & Griesemer 1989), but the research has been limited in its
discussion of design principles that make these boundary objects effective.
Furthermore, the findings from this dissertation study suggest that it may not be the
types of boundary objects per se that are important, but rather the design principles
behind them.
For example, findings from this dissertation study suggest that the boundary
objects being used to represent knowledge support the principles of ownership and
multiplicity. They should be designed to make different points of view more visible
and enable individuals to associate these points of view with the people that hold
them. The boundary objects being used to learn about other people’s knowledge
should support the principle of easy travel. They should be designed to allow people
to link knowledge summaries to details (e.g. assumptions, constraints). The
boundary objects being used to transform knowledge should support the principles of
indeterminacy and emergence. They should be designed such that people are able to
actively add to the knowledge as well as capture and track the additions that evolve
over time and across contexts.
The findings from this dissertation study also contribute to the discussion
about the value of technology vs. people-to-people interaction when it comes to
bridging the barriers to cross-departmental knowledge access. There is much prior
research to suggest that people-to-people interaction should be the primary means of
support (e.g. Dyer & Noveoka, 2000; Hansen et al., 1999; Hargadon & Sutton, 1997;
Nahapiet & Ghoshal, 1998; Nonaka & Takeuchi, 1995; O'Dell & Grayson, 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 5
However, unlike this dissertation study, much of this research has not examined the
use of technology features and people-to-people interaction simultaneously in the
same study. This dissertation study found the use of supportive technology features
can positively impact the perceived value of cross-departmental knowledge on
innovation. Similarly, in another study innovators were found to use technology and
people-to-people interaction to search, evaluate, and integrate other people’s
knowledge for reuse (Majchrzak et ah, 2004). The authors found no evidence that
people-to-people interaction was used more than or instead of technology. These
findings support recent research that has suggested further examination of
complementary as opposed to competing relationships between technology and
people-to-people interaction (Olivera, 2000).
6.3 Future Research
The findings from this dissertation study also yield several interesting ideas
for future research. For example, it was interesting to find that many different KM
technologies were employed to facilitate the use of technology features that support
cross-departmental understanding. In addition, many of them were relatively low-
tech KM technologies that were easy to use. It would be interesting to determine
whether one KM technology with an integrated set of technology features would be
better for innovation than a non-integrated best of breed KM technology solution. It
would also be interesting to compare the perceived ease of use of one integrated KM
technology vs. many non-integrated KM technologies and determine whether the
differences impact the perceived value of cross-departmental knowledge on
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 6
innovation. One the one hand the former may be perceived less easy to use, because
in all likelihood it would have a more complex design. On the other hand, the
former may be perceived to be easier to use because it offers one stop for all
knowledge access.
Future research should also consider the difference between too much and not
enough contextual detail when it comes to information overload and innovation.
Conventional wisdom would say individuals accessing cross-departmental
knowledge value general abstract knowledge, because it is less likely to lead to
information overload and easier to transfer. However, contrary to conventional
wisdom, findings from this dissertation study suggest that individuals accessing
cross-departmental knowledge value general abstract knowledge, only when it is
accompanied by additional details. Future research might investigate how
individuals deal with the information overload expected to result from accessing
cross-departmental knowledge at various levels of detail and consider how
technology might be designed to help.
Follow on research might also consider whether self-managed vs.
technology-managed knowledge access makes a difference with respect to their
impact on the perceived value of cross-departmental knowledge on innovation. For
example, individuals who depend on cross-departmental knowledge to help solve
their problems have to do more than gather the diverse knowledge. They also have
to compare and contrast the alternative perspectives that result and synthesize these
perspectives with their own knowledge to create a solution that meets their needs
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 7
(e.g. Hargadon and Sutton; Majchrzak et al.). Performing these activities through a
self-managed process takes a lot of time and effort. Future research might consider
the extent to which some of these activities can be off-loaded onto KM technologies
and if so, what impact the off-loading would have on individuals’ perceived value of
cross-departmental knowledge on innovation.
It would also be interesting future research to determine whether individuals
using technology features that support cross-departmental understanding experience
intellectual development in these other areas of expertise. Prior research suggests
that individuals develop several different levels of intellect as they become more
skilled in their own area of expertise (Quinn et al., 1996). Individuals develop know-
what, which is generalized knowledge that shares a common syntax and meaning
among individuals (Kogut & Zander, 1992; Nonaka & Takeuchi, 1995). They also
develop know-how, which represents skilled expertise (Kogut & Zander, 1992) and
know-why, which describes the cause and effect relationships that underlie
knowledge (Alavi & Leidner, 2001; Quinn et al., 1996). Research suggests that
individuals develop these different levels of intellect based on their time and
experience within the area of expertise (Nonaka & Takeuchi, 1995; Quinn et al.,
1996). It would be interesting future research to determine whether individuals
could use technology features that support cross-departmental understanding to
develop different levels of intellect in other areas of expertise and if so, whether the
development of intellect might occur at a faster pace given the use of supportive KM
technologies.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 8
Future research would also do well to conduct additional studies where more
variability can be expected in other factors in work environment, such as perceived
openness of communication, perceived task complexity, perceived ease of use of
technology. This dissertation study took place in an organization that provided a
supportive environment for individuals to access cross-departmental knowledge.
Management was actively working to provide continued organizational and
technological support to enable employees to access each other’s knowledge. In
addition, findings indicated that most employees agreed that their work was
complex, the communication across departments was open and KM technologies
were easy to use. Under these circumstances, technology features that support
cross-departmental understanding may have served to complement these other
factors in the work environment and as a result may have been better received. It
would be interesting future research to consider whether technology features that
support cross-departmental understanding can compensate for less supportive
environments. Future research might investigate whether the use of technology
features that support cross-departmental understanding helps individuals overcome
the not invented here syndrome. Future research that determines the extent to which
the other factors in the work environment are baseline conditions for using the
technology features that support cross-departmental understanding would also be a
fruitful area of study. Future research should also reexamine the moderating effects
the other factors in the work environment might have on the relationship between the
use of supportive technology features and the perceived value of cross-departmental
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 9
knowledge on innovation in environments were more variability of the other factors
can be expected.
6.4 Contributions to Practice
Organizations wanting to influence employee innovation are challenged to
get employees to seek, share, or otherwise transfer knowledge across organizational
boundaries. Findings from this research study suggest that the use of supportive
technology features positively influence individuals’ perceived value of cross-
departmental knowledge on innovation. More specifically, the research implies that
organizations should carefully evaluate technology features to determine whether
they support cross-departmental understanding. For example, providing technology
features that allow for periodic updates to knowledge may not be sufficient.
Findings from this dissertation study suggest implementing technology features that
support timely, personalized signals of knowledge change that result from its
ongoing evolution are more likely to be beneficial. The findings also suggest that
technology features should not support the sanitization of knowledge to remove
biases and inconsistencies. Instead, it may be more helpful if technology features
retain the contradictions that result from the variety of personal views and opinions,
and provide a means to facilitate the identification and resolution of the
contradictions. The findings also suggest that technology features not be used to
reduce information overload. Rather, technology features should provide a means to
manage the relationships that result from capturing knowledge at various levels of
detail.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
6.6 Conclusion
This dissertation study extends the IS literature’s examination of the design,
use, and success of technology by examining supportive technology features that
help bridge cross-departmental knowledge barriers. More specifically, a model
proposed that the use of technology features that support cross-departmental
understanding would be positively associated with individuals’ perceived value of
cross-departmental knowledge on innovation. Results from a two-wave survey
supported this assertion. Findings indicated that individuals who use technology
features to help find diverse opinions, see how knowledge about a situation has
evolved over time, link general overviews of knowledge about a situation to
additional related details, and actively add to the knowledge as it evolves, also
believe cross-departmental knowledge helps their divergent thinking and convergent
thinking, two recognized components of innovation. The results from this
dissertation study hold promise for further examination of related phenomena and
can be used to begin to build a cumulative body of IS research in the area.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
121
REFERENCES
Aiken, L. S., & West, S. G. (1991). Multiple Regression: Testing and
Interpreting Interactions. Newbury Park, CA: Sage Publications, Inc.
Alavi, M., & Leidner, D. E. (2001). Review: Knowledge management and
knowledge management systems: Conceptual foundations and research issues. MIS
Quarterly, 25(1), 107-136.
Amabile, T. M. (1988). A model of creativity and innovation in
organizations. In B. M. Staw & L. L. Cummings (Eds.), Research in Organizational
Behavior (Vol. 10, pp. 123-167). Greenwich, CT: JAI Press Inc.
Amabile, T. M., Conti, R., Coon, H., Lazenby, J., & Herron, M. (1996).
Assessing the work environment for creativity. Academy o f Management Journal,
39(5), 1154-1184.
Andrews, K. M., & Delahaye, B. L. (2000). Influences on knowledge
processes in organizational learning: The psychosocial filter. Journal o f
Management Studies, 37(6), 797-810.
Argote, L., & Ingram, P. (2000). Knowledge transfer: A basis for competitive
advantage in firms. Organizational Behavior and Human Decision Processes, <52(1),
150-169.
Basadur, M. (1994). Managing the creative process in organizations. In M. A.
Runco (Ed.), Problem Finding, Problem Solving, and Creativity (pp. 237-268).
Norwood, New Jersey: Ablex Publishing Corporation.
Becerra-Femandez, I., & Sabherwal, R. (2001). Organizational knowledge
management: A contingency perspective. Journal o f Management Information
Systems, 18(1), 23-55.
Benbasat, I., & Lim, L. (1993). The effects of group, task, context, and
technology variables on the usefulness of group support systems: A meta-analysis of
experimental studies. Small Group Research, 24(4), 430-462.
Benbasat, I., & Zmud, R. W. (2003). The identity crisis within the IS
discipline: Defining and communicating the discipline's core properties. MIS
Quarterly, 27(2), 183-194.
Boland, R. J., & Tenkasi, R. V. (1995). Perspective making and perspective
taking in communities of knowing. Organization Science, 6(4), 350-372.
Boland, R. J., Tenkasi, R. V., & Te'eni, D. (1994). Designing information
technology to support distributed cognition. Organization Science, 5(3), 456-475.
Borgatti, S. P., & Cross, R. (2003). A relational view of information seeking
and learning in social networks. Management Science, 49(4), 432-445.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
122
Brophy, D. R. (1998). Understanding, measuring, and enhancing individual
creative problem-solving efforts. Creativity Research Journal, 11(2), 123-150.
Brown, J. S., & Duguid, P. (1991). Organizational learning and communities-
of-practice: Toward a unified view of working, learning, and innovation.
Organization Science, 2(1), 40-57.
Carlile, P. R. (2002). A pragmatic view of knowledge and boundaries:
Boundary objects in new product development. Organization Science, 13(4), 442-
455.
Chin, W. C. (1998b). The partial least squares approach for structural
equation modeling. In G. A. Marcoulides (Ed.), M odem Methods fo r Business
Research (pp. 295-336): Lawrence Erlbaum Associates.
Cohen, J. (1988). Statistical Power Analysis fo r the Behavioral Sciences.
Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Couger, J. D. (1996). Creativity & Innovation in Information Systems
Organizations: International Thompson.
Davis, F. (1989). Perceived usefulness, perceived ease of use, and user
acceptance of information technology. MIS Quarterly, 13(3), 319-340.
Dixon, N. (2000). Common Knowledge. Boston: Harvard Business School
Press.
Dougherty, D. (1992). Interpretive barriers to successful product innovation
in large firms. Organization Science, 3, 179-202.
Dyer, J. H., & Noveoka, K. (2000). Creating and managing a high-
performing knowledge-sharing network: The Toyota case. Strategic Management
Journal, 21, 345-367.
El Sawy, O. A., & Bowles, G. (1997). Redesigning the customer support
process for the electronic economy: Insights from storage dimensions. MIS
Quarterly, 21(4), 457-483.
El Sherif, H., & El Sawy, O. A. (1988). Issue-based decision support systems
for the Egyptian cabinet. MIS Quarterly, 551-569.
Faniel, I. M., & Majchrzak, A. (2002, August 8-11, 2002). An exploratory
study o f the factors associated with the successful reuse o f other's knowledge. Paper
presented at the Eighth Americas Conference on Information Systems, Dallas, Texas.
Faniel, I. M., & Majchrzak, A. (2003, March 14-15). Designing IT for
knowledge reuse in a complex and uncertain environment. Minnesota Symposium on
Knowledge Management.
Filion, F. L. (1975-1976). Estimating bias due to nonresponse in main
surveys. Public Opinion Quarterly, 40, 482-492.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
Garfield, M. J., Taylor, N. J., Dennis, A. R., & Satzinger, J. W. (2001).
Research report: Modifying paradigms - Individual differences, creativity
techniques, and exposure to ideas in group idea generation. Information Systems
Research, 22(3), 322-333.
George, J. M., & Zhou, J. (2001). When openness to experience and
conscientiousness are related to creative behavior: An interactional approach.
Journal o f Applied Psychology, 86(3), 513-524.
Goodman, P. S., & Darr, E. D. (1998). Computer-aided systems and
communities: Mechanisms for organizational learning in distributed environments.
MIS Quarterly, 22(4), 417-440.
Grant, R. M. (1996). Toward a knowledge-based theory of the firm. Strategic
Management Journal, 17, 109-122.
Gray, P. H. (2000). The effects of knowledge management systems on
emergent teams: Towards a research model. Journal o f Strategic Information
Systems, 9, 175-191.
Hair Jr., J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998).
Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall.
Hansen, M. T. (1999). The search-transfer problem: The role of weak ties in
sharing knowledge across organization subunits. Administrative Science Quarterly,
44, 82-111.
Hansen, M. T., Nohria, N., & Tiemey, T. (1999). What's your strategy for
managing knowledge? Harvard Business Review, 77(2), 106-116.
Hargadon, A., & Sutton, R. I. (1997). Technology brokering and innovation
in a product development firm. Administrative Science Quarterly, 42, 716-749.
Henderson, K. (1991). Flexible sketches and inflexible data bases: Visual
communication, conscription devices, and boundary objects in design engineering.
Science, Technology, & Human Values, 76(4), 448-473.
Hirschheim, R., & Klein, H. K. (1994). Realizing emancipatory principles in
information systems development: The case for ETHICS. MIS Quarterly, 83-109.
Huber, G. P. (1990). A theory of the effects on advanced information
technologies on organizational design, intelligence, and decision making. Academy
o f Management Review, 75(1), 47-71.
Jabri, M. M. (1991). The development of conceptually independent subscales
in the measurement of modes of problem solving. Educational and Psychological
Measurement, 51, 975-983.
Kanter, R. M. (1988). When a thousand flowers bloom: Structural,
collective, and social conditions for innovation in organization. In B. M. Staw & L.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 4
L. Cummings (Eds.), Research in Organizational Behavior (Vol. 10, pp. 169-211).
Greenwich, CT: JAI Press Inc.
Kogut, B., & Zander, U. (1992). Knowledge of the firm, combinative
capabilities, and the replication of technology. Organization Science, 3(3), 383-397.
Lave, J., & Wenger, E. (1991). Situated Learning: Legitimate Peripheral
Participation. Cambridge: Cambridge University Press.
Leonard, D., & Sensiper, S. (1998). The role of tacit knowledge in group
innovation. California Management Review, 40(3), 112-132.
Majchrzak, A., Cooper, L. M., & Neece, O. (2004). Knowledge reuse for
innovation. Management Science, 50(2), 174-188.
Markus, M. L. (2001). Toward a theory of knowledge reuse: Types of
knowledge reuse situations and factors in reuse success. Journal o f Management
Information Systems, 18(1), 57-93.
Markus, M. L., Majchrzak, A., & Gasser, L. (2002). A design theory for
systems that support emergent knowledge processes. MIS Quarterly, 26(3), 199-232.
Mason, R. O., & Mitroff, 1.1. (1973). A program for research on management
information systems. Management Science, 19(5), 475-487.
Mayer, C. S., & Pratt, J., R.W. (1966). A note on nonresponse in a mail
survey. Public Opinion Quarterly, 30, 667-646.
Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the
organizational advantage. Academy o f Management Journal, 23(2), 242-266.
Nohria, N„ & Eccles, R. G. (1992). Face-to-Face: Making network
organizations work. In N. Nohria & R. G. Eccles (Eds.), Networks and
Organizations: Structure, Form, and Action (pp. 288-308). Boston: Harvard
Business School Press.
Nonaka, I., & Takeuchi, H. (1995). The Knowledge Creating Company. New
York: Oxford University Press.
Nunamaker Jr., J. F., Briggs, R. O., Mittleman, D. D., Vogel, D. R., &
Balthazard, P. A. (1997). Lessons from a dozen years of group support systems
research: A discussion of lab and field findings. Journal o f Management Information
Systems, 13(3), 163-207.
Nunnally, J. C. (1978). Psychometric Theory. New York: McGraw-Hill.
O'Dell, C., & Grayson, C. J. (1998). If only we know what we know:
Identification and transfer of internal best practices. California Management Review,
40(3), 154-174.
Oldham, G. R., & Cummings, A. (1996). Employee Creativity: Personal and
contextual factors at work. Academy o f Management Journal, 39(3), 607-634.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
Olivera, F. (2000). Memory systems in organizations: An empirical
investigation of mechanisms for knowledge collection, storage, and access. Journal
o f Management Studies, 37(6), 811-832.
O'Reilly, C. A., & Roberts, K. H. (1977). Task group structure,
communication, and effectiveness in three organizations. Journal o f Applied
Psychology, 62(6), 674-681.
Orlikowski, W. J. (2000). Using technology and constituting structures: A
practice lens for studying technology in organizations. Organization Science, 11(4),
404-428.
Orlikowski, W. J., & Iacono, C. S. (2001). Research Commentary:
Desperately seeking the "IT" in IT research - A call to theorizing the IT artifact.
Information Systems Research, 12(2), 121-134.
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. (2003).
Common method biases in behavioral research: A critical review of the literature
and recommended remedies. Journal o f Applied Psychology, 88(5), 879-903.
Quinn, J. B., Anderson, P., & Finkelstein, S. (1996). Managing professional
intellect: Making the most of the best. Harvard Business Review, 71-80.
Ruggles, R. (1998). The state of the notion: Knowledge management in
practice. California Management Review, 40(3), 80-89.
Runco, M. A. (1994). Conclusions concerning problem finding, problem
solving, and creativity. In M. A. Runco (Ed.), Problem Finding, Problem Solving,
and Creativity (pp. 271-290). Norwood, New Jersey: Ablex Publishing Corporation.
Runco, M. A. (1999). Divergent thinking. In M. A. Runco & S. R. Pritzker
(Eds.), Encyclopedia o f Creativity (Vol. 2, pp. 577-582). San Diego: Academic
Press.
Satzinger, J. W., Garfield, M. J., & Nagasundaram, M. (1999). The creative
process: The effects of group memory on individual idea generation. Journal of
Management Information Systems, 15(4), 143-160.
Scott, S. G., & Bruce, R. A. (1994). Determinants of innovative behavior: A
path model of individual innovation in the workplace. Academy o f Management
Journal, 37(3), 580-607.
Simon, H. A. (1991). Bounded Rationality and Organizational Learning.
Organization Science, 2(1), 125-134.
Sprague Jr., R. H. (1980). A framework for the development of decision
support systems. MIS Quarterly, 1-26.
Star, S. L., & Griesemer, J. R. (1989). Institutional ecology, 'translations' and
boundary objects: Amateurs and professionals in Berkeley's museum of vertebrate
zoology. Social Studies o f Science, 19(3), 387-420.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 6
Stein, E. W., & Zwass, V. (1995). Actualizing organizational memory with
information systems. Information Systems Research, (5(1), 85-117.
Sussman, S. W., & Siegal, W. S. (2003). Informational influence in
organizations: An integrated approach to knowledge adoption. Information Systems
Research, 14( 1), 47-65.
Szulanski, G. (1996). Exploring internal stickiness: Impediments to the
transfer of best practice with the firm. Strategic Management Journal, 77(10), 27-43.
Treffinger, D. J., Isaksen, S. G., & Dorval, K. B. (1994). Creative problem
solving: An overview. In M. A. Runco (Ed.), Problem Finding, Problem Solving,
and Creativity (pp. 223-236). Norwood, New Jersey: Ablex Publishing Corporation.
Vandenbosch, B., & Higgins, C. A. (1995). Executive support systems and
learning: A model and empirical test. Journal o f Management Information Systems,
72(2), 99-130.
von Hippie, E. (1994). "Sticky information" and the locus of problem
solving: Implications for innovation. Management Science, 40(4), 429-439.
von Krogh, G. (1998). Care in knowledge creation. California Management
Review, 40(3), 133-153.
Walsh, J. P., & Ungson, G. R. (1991). Organizational memory. Academy o f
Management Review, 16(1), 57-91.
Yoon, Y., Guimaraes, T., & O'Neal, Q. (1995). Exploring the factors
associated with expert systems success. MIS Quarterly, 79(1), 83-106.
Zigurs, I., & Buckland, B. K. (1998). A theory of task/technology fit and
group support systems effectiveness. MIS Quarterly, 22(3), 313-334.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 7
APPENDIX A
Instructions: When answering the questions in this survey, please think about how you typically access and reuse
knowledge from your Company A co-workers in the course of doing your regular professional work at Company
A.
Please enter your Company A badge number:___________
In the course of doing your regular professional work at Company A, you access knowledge from Company A
co-workers, please use the scale below to indicate the frequency you use the following IT systems to access the
knowledge.
Email:
None
of the
time
1
Some
of the
time
2
Most
of the
time
3
Databases: 1 2 3
Intranet: 1 2 3
Instant messaging, live discussion groups: 1 2 3
Modeling/simulation tools: 1 2 3
Shared file servers: 1 2 3
Knowledge management infrastructure: 1 2 3
Threaded discussions, electronic bulletin boards: 1 2 3
Source code libraries/databases: 1 2 3
Distribution lists, list servers: 1 2 3
In the course of doing your regular professional work at Company A, you access knowledge from Company A
co-workers in other departments/areas, please use the scales below to indicate how frequently the IT systems you
indicated using are used to ...
To no To a To To a To a
extent minor
extent
some
extent
major
extent
great
extent
find the changes made to the knowledge as ideas evolve over
time (e.g. access current and historical program data).
1 2 3 4 5
keep track of how the knowledge changes over time (e.g.
account for design, material, or part changes over life of a
program).
1 2 3 4 5
identify the experts (e.g. authors) of the knowledge (e.g.
memos, source code).
1 2 3 4 5
access suggested solutions, thoughts, opinions, and/or analyses
(e.g. future predictions on part reliability, trend analyses).
1 2 3 4 5
find others with the experience you need (e.g. specific skills,
program insights).
1 2 3 4 5
find knowledge that has been contributed by specific
individuals.
1 2 3 4 5
identify multiple ways to approach your problem/task (e.g.
competing theories, designs, algorithms).
1 2 3 4 5
access knowledge that offers different points of view (e.g.
comparisons between parts, materials, or designs).
1 2 3 4 5
be informed when knowledge you have an interest in changes
(e.g. reports of latest failures/anomalies).
1 2 3 4 5
find alternative ideas (e.g. compare output from or perform
what if scenarios).
1 2 3 4 5
find additional knowledge that shares a similar subject or
purpose (e.g. sources referenced in memos, reports, briefings).
1 2 3 4 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 8
To no To a To To a To a
extent minor some major great
access details that put the knowledge into context (e.g. 1
extent
2
extent
3
extent
4
extent
5
technical specifications, input parameters).
find know ledge that supports the rationale behind the decisions 1 2 3 4 5
made by other people so that the decisions and rationale can be
revisited later (e.g. documented assumptions, constraints,
decision trees).
find summaries (e.g. technical narratives, design reviews) as 1 2 3 4 5
well as details (e.g. test data, source code, technical
configurations) about the knowledge.
find associated material that supports the knowledge (e.g. 1 2 3 4 5
notes, email, raw data, or schematics).
Please use the scale below to indicate the extent you agree with the following statements about the IT Systems
you use when accessing knowledge from your co-workers at Company A.
Strongly
disagree
Disagree Neutral Agree Strongly
agree
Learning to operate the IT systems is
easy for me.
1 2 3 4 5
I find it easy to get the IT systems to do
what I want them to do.
1 2 3 4 5
My interaction with the IT systems is
clear and understandable.
1 2 3 4 5
I find the IT systems flexible to interact
with.
1 2 3 4 5
It was easy for me to become skillful at
using the IT systems.
1 2 3 4 5
I find the IT systems easy to use. 1 2 3 4 5
In the course of doing your regular professional work at Company A, you access knowledge from Company A
co-workers in other departments/areas, please use the scales below to indicate how frequently you use the
following activities to access the knowledge.
To no To a To To a To a
extent minor some major great
extent extent extent extent
Technical seminars 1 2 3 4 5
Common projects worked on in the past (e.g. you and co-worker(s) 1 2 3 4 5
assigned to same project)
Briefings (e.g. verbal presentations) 1 2 3 4 5
Informal peer reviews 1 2 3 4 5
Unscheduled meetings (e.g. cafeteria, hallway, desk) 1 2 3 4 5
Corporate events (e.g. CEO’s report to employees) 1 2 3 4 5
Guest lectures 1 2 3 4 5
Company A Institute courses 1 2 3 4 5
Telephone conversations 1 2 3 4 5
Video conferences 1 2 3 4 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 9
In the course of doing your regular professional work at Company A, indicate the extent accessing knowledge
from Company A co-workers in other departments/areas helps you...
To no To a To To a To a
extent minor
extent
some
extent
major
extent
great
extent
consider various problem solving ideas that may or may not lead
you to a solution.
1 2 3 4 5
adhere to the commonly established rules of your area of work. 1 2 3 4 5
pursue a problem, particularly if it takes you into areas you don’t
know much about.
1 2 3 4 5
follow generally accepted methods for solving problems. 1 2 3 4 5
search for novel approaches not required at the time. 1 2 3 4 5
Pay strict regard to the sequence of steps needed for the
completion of a job.
1 2 3 4 5
link ideas which stem from more than one area of investigation. 1 2 3 4 5
adhere to the well-known problem solving techniques, methods,
and procedures of your area of work.
1 2 3 4 5
readily accept the usual and generally proven methods of
solution.
1 2 3 4 5
make unusual connections about ideas even if they are trivial. 1 2 3 4 5
be precise and exact about production of results. 1 2 3 4 5
trace relationships between disparate areas of work. 1 2 3 4 5
engage in what appear to be novel methods of solution. 1 2 3 4 5
adhere carefully to the standards of your area of work. 1 2 3 4 5
be fully aware of the sequence of steps required in solving
problems.
1 2 3 4 5
be rigorous in deriving your solutions. 1 2 3 4 5
make connections between apparently unrelated ideas. 1 2 3 4 5
be methodical in the way you tackle the problems. 1 2 3 4 5
think about more than one concept, method, or solution. 1 2 3 4 5
Please indicate the extent you agree with the following statements about communication between you and
Company A co-workers in other departments/areas.
Strongly Disagree
disagree
Neutral Agree Strongly
agree
It is easy to talk openly to employees at
Company A.
Communication at Company A is very
open.
I find it enjoyable to talk to others at
Company A.
When people talk to each other there is a
great deal of understanding.
It is easy to ask advice from any Company
A employee.
2 3 4 5
2 3 4 5
2 3 4 5
2 3 4 5
2 3 4 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 0
In the course of doing your regular professional work at Company A, please indicate the extent you agree with
the following statements.
Often you are unable to anticipate w hat the
end results of your efforts at work will be.
It is often difficult to determine how to
reach work goals or objectives.
Often, there are several different ways in
which to perform this job.
This job often requires you to take time to
figure out what you should be trying to
accomplish.
You often need to think about how to go
about performing this job.
Many of the tasks on this job can be
performed in different ways.
In starting a project or task, it is not always
clear what exactly should be achieved.
You often need to choose how to perform
work tasks on this job.
More often than not, there are multiple
ways to achieve work goals or objectives.
Sometimes you need to change goals or
objectives as you get involved in a work
project or task.
Figuring out how to achieve work goals or
objectives is often a real challenge.
Deciding which way to proceed on a work
task or project is often a challenge.
The results to be accomplished on this job
are relatively clear cut.
There is often one best way to perform this
job.
Strongly Disagree Neutral Agree Strongly
Disagree agree
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Years at Company A: ___ 0 to 5 ____6 to 10 ___ 11 to 15 ___ 16 to 20 ___ 21 or more
Division: V
Subdivision: Department:,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
APPENDIX B
Instructions: When answering the questions in this survey, please think about how you typically access and reuse
knowledge from your Company A co-workers in the course of doing your regular professional work at Company
A.
Please enter your Company A badge number:___________
In the course o f doing your regular professional work at Company A, indicate the extent accessing knowledge
from Company A co-workers in other departments/areas helps you...
To no To a To To a To a
extent minor some major great
extent extent extent extent
consider various problem solving ideas that may or may not lead 1 2 3 4 5
you to a solution.
adhere to the commonly established rules of your area of work. 1 2 3 4 5
pursue a problem, particularly if it takes you into areas you don’t 1 2 3 4 5
know much about.
follow generally accepted methods for solving problems. 1 2 3 4 5
search for novel approaches not required at the time. 1 2 3 4 5
Pay strict regard to the sequence of steps needed for the 1 2 3 4 5
completion of a job.
link ideas which stem from more than one area of investigation. 1 2 3 4 5
adhere to the well-known problem solving techniques, methods, 1 2 3 4 5
and procedures of your area of work.
readily accept the usual and generally proven methods of 1 2 3 4 5
solution.
make unusual connections about ideas even if they are trivial. 1 2 3 4 5
be precise and exact about production of results. 1 2 3 4 5
trace relationships between disparate areas of work. 1 2 3 4 5
engage in what appear to be novel methods of solution. 1 2 3 4 5
adhere carefully to the standards of your area of work. 1 2 3 4 5
be fully aware of the sequence of steps required in solving 1 2 3 4 5
problems.
be rigorous in deriving your solutions. 1 2 3 4 5
make connections between apparently unrelated ideas. 1 2 3 4 5
be methodical in the way you tackle the problems. 1 2 3 4 5
think about more than one concept, method, or solution. 1 2 3 4 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Essays on technological evolution and financial returns to innovation

PDF

Defining the advice -seeking request in electronic solicitations in boundary spanning environments

PDF

Information systems capabilities and indirect relational value: Sustaining networked organizations

PDF

Essays on technology innovation and new product development strategy

PDF

Board involvement in monitoring and strategy making: Antecedents and consequences

PDF

Essays in supply chain management

PDF

Demographic diversity, team process, and team performance: Assessing moderator effects of cognitive conflict management practices and task interdependence

PDF

Inventory and pricing models for perishable products

PDF

Corporate ethical orientation and crisis management before and after September 11, 2001

PDF

Does adding nonfinancial value drivers to a summary financial measure improve the learning and performance of managers?

PDF

Effects of redundancy in media coverage on nonprofessional investors' earnings forecasts

PDF

Biopharmaceutical strategic alliances: Interorganizational dynamics and factors influencing FDA regulatory outcomes

PDF

Essays on new product preannouncements

PDF

Consumer choice under budget constraint: Why consumers overspend

PDF

Feeling good about control: Design considerations for accountability systems in schools

PDF

Androgyny and transformational leadership: Effects of gender and sex -role identity in the collectivistic context of Taiwan, R.O.C.

PDF

A typology of corporate environemental strategy and its driving factors in jultinational corporations

PDF

Customer satisfaction, customer bargaining power, and financial performance

PDF

An investigation of repeat visit behavior on the Internet: Models and applications

PDF

Information diffusion and the boundary of market efficiency

Asset Metadata

Creator Faniel, Ixchel Marika (author)

Core Title Influencing individual innovation through technology features that support cross -departmental understanding

School Graduate School

Degree Doctor of Philosophy

Degree Program Business Administration

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag business administration, management,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Majchrzak, Ann (committee chair), James, Gareth (committee member), Josefek, Robert (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-568517

Unique identifier UC11335854

Identifier 3155406.pdf (filename),usctheses-c16-568517 (legacy record id)

Legacy Identifier 3155406.pdf

Dmrecord 568517

Document Type Dissertation

Rights Faniel, Ixchel Marika

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA