Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Defining the advice -seeking request in electronic solicitations in boundary spanning environments
(USC Thesis Other)
Defining the advice -seeking request in electronic solicitations in boundary spanning environments
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
NOTE TO USERS
This reproduction is the best copy available.
®
UMI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEFINING THE ADVICE-SEEKING REQUEST IN ELECTRONIC
SOLICITATIONS IN BOUNDARY SPANNING ENVIRONMENTS
Copyright 2005
by
Jiangfan Jenny Zhong
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)
August 2005
Jiangfan Jenny Zhong
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3196926
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 3196926
Copyright 2006 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
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
To my dearest family
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
ACKNOWLEDGEMENTS
First, I would like to thank my parents who brought me to this world and
God for giving me both the health and strength all these years. I thank my
advisor, Prof. Ann Majchrzak, for her meticulous feedback, helpful guidance, and
precious experience. This work would not have been possible without her
support. I am also very thankful to my committee members and qual committees,
Prof. Omar El Sawy, Prof. Janet Fulk, Prof. Gareth James, Prof. Ronald Rice,
Prof. Robert Jesefek, and Prof. Ram Chellappa for their priceless feedback and
words of encouragements.
Thank you to Dr. Ann Majchrzak for guiding this dissertation. Her
guidance made this process all the more worthwhile. I am thankful for the
extensive amounts of time she spent on helping refine key points of this
dissertation. I am thankful to Dr. Omar El Sawy for his constant prodding to
make my research relevant and interesting. I also thank him for this continuous
encouragement and attention to my research progress. I also thank Dr. Janet Fulk
for her willingness to help me solve the research issues that I faced in the early
stages. I am thankful to Dr. Ronald Rice for being willing and able to help
remotely. I appreciate his guidance in helping me extract the learning from the
study that actually would make this process meaningful. I am thankful to Dr.
Gareth James for helping me solve the methodology issues. I thank Dr. Robert
Josefek for giving me the opportunity to see how excellent teaching should be
done. I am also thankful to Dr. Lucy Lee who helped me conquer the
communication barriers in the early stage of my study. I also want to thank
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
Nancy Stoffer, who always welcomed me with her smile when I stopped by her
office with questions. Thank you to Deborah, Elizabeth, and Tina who constantly
provide the best support. Thank you to IS students past and present, Ricky,
Nilesh, Ixchel, Paul, Shaosong, Raymond, and Dave who helped me during my
study at the Marshall School of Business.
I am also grateful and have much love for my family and friends. I have
too many to name individually. Words cannot express my gratitude. I just want to
thank all my families for their love, constant support, and encouragement across
time and space - which makes the completion of my study in another continent
possible. I could not have asked for better husband, who has been so
understanding, encouraging, and supportive.
Many thanks to Dr. Samer Faraj, for his help in connecting me with the
site company for my data collection - without this help my research would not
have steered to completion. Finally, special thanks to my sponsor, Chemonics
International Company, for giving me the opportunity to be enrolled in the
company’s e-mail account; and to the University of Southern California in
admitting me into her great graduate program.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
TABLE OF CONTENTS
D E D IC A T IO N ....................................................................................................................................... II
A C K N O W L E D G E M E N T S.............................................................................................................Ill
LIST OF T A B L E S ........................................................................................................................... VIII
LIST OF F IG U R E S ............................................................................................................................IX
A B S T R A C T ............................................................................................................................................X
CHAPTER 1: EX EC U TIV E S U M M A R Y ...................................................................................1
1.1 M o tiv a tio n................................................................................................................................. 2
1.2 Overview of C onceptual M odel a n d H y p o t h e s e s.............................................. 5
1.3 O verview of R esearch M e t h o d o lo g y........................................................................ 8
1.4 Su m m a r y of Fin d in g s.......................................................................................................... 11
1.5 Su m m a r y of C o n t r ib u t io n s............................................................................................12
1.6 R o adm ap to T he D isse r t a t io n.......................................................................................17
CHAPTER 2: IN T R O D U C T IO N .................................................................................................. 18
2.1 M otivation for St u d y ......................................................................................................21
2.2 Electronic Kno w ledg e T r a n sfe r...............................................................................25
2.3 Kn o w led g e Tra n sfer A cross B o u n d a r ie s............................................................ 29
2.4 How to F acilitate A chievem ent of In t e n d e d L e a r n in g b y B ridging
B o u n d a r y B a r r ie r s...................................................................................................................31
2.5 Su m m a r y ....................................................................................................................................32
CHAPTER 3: LITERATURE RESEARCH A N D C O N C EPTU A L M O D E L 35
3.1 Theory of T hought W o r l d s.......................................................................................... 36
3.2 Th eory of Kno w ledg e R e u se......................................................................................... 39
3.3 Im portance of M eeting L earning In t e n t io n s......................................................42
3.3.1 Achieving Intended Learning in a Cross-Boundary Context.........................43
3.3.2 Types o f Learning Intentions.....................................................................................47
3.4 D efining the A d v ic e-S eeking R equest to F acilitate A chievem ent of
Int e n d e d L e a r n in g ..................................................................................................................... 48
3.4.1 Richness of Situational Context................................................................................. 49
3.4.1.1 Importance o f Specifying Situational C on text.............................................. 50
3.4.1.2 E ffect o f Situational C ontext................................................................................53
3.4.2 Explication of Learning Intent..................................................................................... 55
3.4.2.1 Importance o f Explication o f Learning Intent.............................................55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vi
3.4.2.2 Effect of Explication of Learning Intent..................................................57
3.4.3 Emotional Incentive........................................................................................................ 59
3.4.3.1 Importance of Providing Emotional Incentive......................................... 60
3.4.3.2 Effect of Emotional Incentive..................................................................... 61
3.4.4 Effect o f Boundary............................................................................................................. 63
3.5 C ontrol V a r ia b l e s.............................................................................................................64
3.6 C o nceptual M o d e l.............................................................................................................. 66
3.7 Su m m a r y ................................................................................................................................... 67
CHAPTER 4: RESEARCH METHODS.........................................................................69
4.1 R esear ch S e t t in g .................................................................................................................70
4.2 Qua lita tiv e St u d y .............................................................................................................. 71
4.3 Q u a n tita tiv e St u d y ............................................................................................................74
4.3.1 Survey Methodology..........................................................................................75
4.3.1.1 Survey Administration...............................................................................76
4.3.1.2 Survey Instruments.................................................................................... 78
4.3.2 Coding Methodology....................................................................................................... 81
4.3.3 Boundary Information.................................................................................................... 84
4.4 D ata A nalytic T e c h n iq u e s.............................................................................................85
CHAPTER 5: DATA ANALYSIS AND RESULTS....................................................88
5.1 D escriptive St a t ist ic s.......................................................................................................90
5.2 D iscrim inant V a l id it y.......................................................................................................91
5.3 Structur al M o d e l ..............................................................................................................95
5.3.1 Testing Proposed M odel.................................................................................. 95
5.3.2 Testing Alternate Structural Paths..................................................................98
CHAPTER 6: DISCUSSION............................................................................................99
6.1 D isc u ssio n Of The R e su l t s..............................................................................................99
6.2 Lim it a t io n s............................................................................................................................102
6.2.1 A Single Site for Data Collection.............................................................................102
6.2.2 Perspectives from Knowledge Seeker....................................................................103
6.2.3 Cross-sectional Data Used........................................................................................104
6.3 C o ntributio ns to T h e o r y.............................................................................................. 105
6.4 Fu tu r e R e s e a r c h ................................................................................................................108
6.5 C ontributions to Pr a c t ic e.......................................................................................... 110
6.6 Co n c l u sio n ............................................................................................................................112
REFERENCES..................................................................................................................114
APPENDICES
APPENDIX A ...................................................................................................................121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
APPENDIX C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 4.1 Examples of Advice-seeking E-mail.............................................................. 71
Table 4.2 An Example of a Well-defined Advice-seeking Request with Proper
Dimensions................................................................................................................. 74
Table 4.3 Examples of Advice Received by E-mail.......................................................76
Table 4.4 List of items measuring the constructs and their Cronbach’s alpha............80
Table 4.5 Coding Procedure and Instructions.................................................................83
Table 4.6 Coding Examples..............................................................................................84
Table 5.1 Summary of Elypotheses...................................................................................89
Table 5.2 Descriptive Statistics........................................................................................ 90
Table 5.3 PLS Component-based Analysis: Cross-loadings.........................................91
Table 5.4 Inter-Construct Correlations............................................................................94
Table 6.1 Summary of the Results..................................................................................100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 1.1 Conceptual Model.............................................................................................. 8
Figure 2.1 Sussman and Siegal’s M odel......................................................................... 29
Figure 3.1 The Hypotheses and Proposed Model ........................................................ 66
Figure 5.1 Structural Model (Associated with Model Maintenance Learning) ....... 96
Figure 5.2 Structural Model (Associated with Model Building Learning) ...............97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X
ABSTRACT
Global knowledge work is becoming increasingly reliant on online
knowledge transfer. This increases the need for employees to access knowledge
from different areas of expertise through electronic distribution lists within the
organization - accessing knowledge from colleagues in other departments.
However, employees face a major challenge in bridging barriers to
cross-boundary knowledge access, barriers that lower the possibility of receiving
useful knowledge. Current theories of e-mail-facilitated knowledge transfer
ignore these barriers by emphasizing the role of argument quality in determining
a knowledge seeker’s perceived usefulness of the knowledge received, assuming
knowledge holders share the same thought worlds with seekers and thus
understand knowledge seekers’ learning needs. This dissertation couples the
theory of thought worlds with Markus’ theory of knowledge reuse to propose
that, when crossing thought worlds, knowledge seekers have different learning
intentions, and that the effect of argument quality on perceived knowledge
usefulness is mediated by whether the provided information meets the knowledge
seekers’ learning intentions. The model also proposes that the way
advice-seeking requests are phrased in e-mail solicitations to cross thought world
barriers contributes to the achievement of intended learning. Three dimensions of
advice-seeking requests are suggested, drawn from Markus’ theory of knowledge
reuse: situational context, explication of learning intent, and emotional incentive.
The research began with a qualitative study including on-site interviews,
observations, and reviews of e-mail examples in an international consulting firm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
xi
This was followed by a two-phase survey of 136 employees of the firm where
knowledge seekers were asked about the advice received from knowledge
holders; and the knowledge seekers’ e-mail solicitations were coded according to
the three dimensions of advice-seeking definition. It was found that in the process
of cross-boundary knowledge access through e-mail distribution lists, the
relationship between knowledge seekers’ perceptions of the argument quality of
the advice received and perceptions of the usefulness of that received knowledge
was mediated by achievement of the seekers’ intended mental model learning.
In addition, two of the three dimensions of advice-seeking definition were
supported: situational context and explication of learning intent. The study
provides theoretical and practical implications. Limitations and future research
are also discussed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER 1: EXECUTIVE SUMMARY
Global knowledge work is becoming increasingly reliant on online
knowledge transfer (Sussman & Siegal, 2003). The key m edia for language
activities involved in knowledge sharing has been evolving from oral communication
to computer-mediated communication (CMC). As such, it has becom e increasingly
important to identify ways to facilitate knowledge transfer in technology-mediated
settings. “Knowledge intensive firms are composed o f multiple communities with
highly specialized technologies and knowledge domains” (Boland & Tenkasi, 1999,
p. 327). W ithin organizations, this often means acquiring useful knowledge from
colleagues in other departments through electronic distribution lists (i.e., different
online communities). However, a major challenge organizations face is bridging the
barriers to cross-departmental knowledge seeking. In order to facilitate
cross-boundary knowledge transfer through CMC, it is suggested in this study that
achieving the intended mental model development is critical. Only when the
intended mental model is achieved can the received knowledge be perceived as
useful. O f particular interest to this dissertation is the impact on achieving intended
mental model learning from communicated messages with well-defined search
questions, using CMC in distributed networks.
In this executive summary, the motivation for this dissertation study is
highlighted (section 1.1). This includes a discussion about the gap in the existing
research, and the value of a well-defined advice-seeking request in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
technology-mediated learning in boundary spanning environments. Next, the
conceptual model and research hypotheses are outlined (section 1.2). This is
followed by a summary o f the research methodology employed in this dissertation
study (section 1.3), findings (section 1.4), and contributions (section 1.5). The
executive summary ends w ith an outline o f the remaining chapters (section 1.6).
1.1 M otivation
Today’s fast-changing and global business environment has made it
necessary to expand the geographical scope o f knowledge workers. Advances in
communication and information technologies, especially the ubiquitous diffusion of
Internet technology, have made this possible (Brown, 1998). “Organizations are
characterized by a process o f distributed cognition in which multiple communities of
specialized knowledge workers, each dealing with a part o f an overall organizational
problem” (Boland & Tenkasi, 1999, p. 329). Computer networks facilitate
employees to access knowledge from different online communities across boundaries
through electronic distribution lists (Constant et al., 1999). “They also make it
possible to ask strangers for advice” (Constant et al., 1999, p. 415). For instance, a
knowledge worker can post a query to a large electronic distribution list without
knowing who might read it (Constant et al., 1999). Nonetheless, knowledge access
crossing organizational boundaries is difficult due to thought world barriers
(Dougherty, 1992).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
Thought world barriers occur owing to different interpretive schemes as
knowledge workers from diverse online communities hold different expertise
(Dougherty, 1992). Members o f online communities share values, norms, and
perspectives, and are connected electronically through the Intranet (i.e., electronic
distribution list). Knowledge workers belonging to different e-mail distribution lists
(i.e., online communities) often hold various thought worlds. These barriers require
knowledge workers to spend huge amounts o f time and effort to extract useful
information from large numbers o f e-mails and discussion postings. This is a major
reason why technologies have been unable to create computer-mediated work
environments that are as successful as physic ally-shared environments (Kraut et al.,
2001).
IS scholars have found some factors to be predictive o f successful knowledge
transfer using CMC, such as expertise and involvement o f knowledge seekers
(Stamm & Dube, 1994; Sussman & Siegal, 2003); expertise o f knowledge holders
(Faraj & Sproull, 2000); credibility o f knowledge holders (Sussman & Siegal, 2003;
W u & Shaffer, 1987); characteristics o f the communication medium (Lee, 1991;
Markus, 1994); argument quality o f the knowledge received (Sussman & Siegal,
2003), and properties o f knowledge (e.g., Nonaka et. al., 2000). Considered as a
whole, the existing research provides a model o f factors that influence knowledge
usefulness. However, current literature on electronic knowledge transfer ignores
whether the advice comes from the same or a different thought world (e.g., Sussman
& Siegal, 2003), thus assuming that accessing knowledge across thought worlds is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
the same as knowledge access from the same thought world. Literature on thought
worlds indicates different cognitive processes are involved to seek advice across
thought worlds (Dougherty, 1992). Thus, the fundamental assumption o f the
e-mail-facilitated advice-seeking literature is questionable.
Markus (2001) develops a theory o f knowledge reuse which provides insights
for how to strategically bridge the barriers in cross-boundary knowledge transfer
processes. She suggests that it is important to meet knowledge seekers’ needs and
intentions. The literature o f cognitive learning supplements the theory o f knowledge
reuse to identify two types o f learning intentions (i.e., model building vs. model
maintenance). In order to achieve knowledge seekers’ learning intentions, the key is
to have a well-defined knowledge seeking request (Markus, 2001) to bridge thought
world barriers manifested as boundaries. The challenge then becom es how to define
knowledge-seeking requests to achieve knowledge seekers’ learning intentions, and
therefore induce transfer o f useful knowledge. In this study, the following question is
addressed: During electronic solicitations by knowledge seekers, how should the
advice-seeking request be defined when organizational boundaries are crossed for
the seeker to achieve his/her intended learning and obtain knowledge he or she
judges useful? Advice is generally viewed as “information that communicates an
opinion about what could or should be done about a problem or issue” (Sussman &
Siegal, 2003, p. 49). In this study, an opinion or recommendation that knowledge
seekers receive in response to their queries would be construed as advice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
1.2 Overview o f Conceptual Model and Hypotheses
In answering the research question posed in this dissertation, thought world,
learning, and defining an advice-seeking request need to be conceptualized. The
theory o f knowledge reuse is coupled with the theory o f thought worlds to delineate
a set o f features to be deployed in a well-defined advice-seeking request so as to
facilitate achieving knowledge seekers’ intended learning.
Current information systems (IS) research has found that knowledge seekers
are more likely to perceive the advice (received from knowledge holders) useful
when the holders do not just present the information, but also offer a cogent
argument for its worth, which has been called “argument quality.” (Sussman &
Siegal, 2003). However, knowledge access crossing boundaries is difficult due to
interpretive barriers - thought world barriers (Dougherty, 1992). “A thought world is
a community o f persons engaged in a certain domain activity who have a shared
understanding about that activity” (Dougherty 1992, p. 182). As such, thought world
barriers are created when crossing organizational boundaries to access knowledge.
This dissertation addresses thought world barriers by taking into account the
different organizational boundaries that need to be crossed in e-mail-facilitated
knowledge transfer processes.
Literature on thought worlds indicates different cognitive processes are
involved when seeking advice across thought worlds (Dougherty, 1992). When
crossing boundaries, knowledge holders may not be able to understand knowledge
seekers’ problem or m ay interpret the problem differently, and thus produce
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
knowledge that cannot meet knowledge seekers’ needs. M any knowledge transfer
failures are attributed to unmet knowledge seekers’ needs (Markus, 2001). For
successful knowledge transfer to occur, knowledge seekers must experience
meaningful learning (Sussman & Siegal, 2003). A cognitive learning perspective
suggests that some knowledge seekers m ay need to achieve model building (MB)
learning while others m ay need to achieve model maintenance (MM) learning
(Vandenbosch & Higgins, 1996). The knowledge received will be judged useful only
when their learning needs are met (Markus, 2001). If the knowledge received cannot
help knowledge seekers to achieve what they intend to leam, anxiety may occur
(Wastell, 1999). Therefore, the knowledge seeker may be reluctant to process and
accept the knowledge received and engage in “defensive, anti-learning processes”
(W asted, 1999). Thus, this dissertation proposes that only when the knowledge
received meets the knowledge seekers’ learning needs or objectives - meaning that
they can achieve desired mental model building vs. maintenance (i.e., to achieve
intended learning) - is the seeker likely to perceive the knowledge to be useful. In
short, in the context o f electronic knowledge transfer across boundaries in online
communities, achieving intended learning mediates the relationship between
argument quality and perceived usefulness o f knowledge received.
The challenge then becomes to facilitate achieving knowledge seekers’
intended learning bridging the barriers. The theory o f knowledge reuse goes beyond
the theory o f thought world to suggest the indispensable effect o f defining
advice-seeking requests in electronic solicitations. Further, it provides insights for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
designing the features o f a cogent and concise advice-seeking request. For example,
explication o f MB intent (vs. explication o f M M intent) encourages knowledge
holders to provide different perspectives and facilitates knowledge seekers to achieve
intended model building learning (vs. model maintenance learning). In addition,
factual and interpretive context (vs. factual only context) facilitates the achievement
o f intended model maintenance learning (vs. model building learning). Furthermore,
detailed emotional incentive included in electronic requests is more important for
achieving intended model building learning than for model maintenance learning.
These relationships will be explained in more detail in Chapter 3.
In short, knowledge workers who access cross-boundary knowledge (in
other online communities) electronically encounter interpretive barriers. Intended
mental model learning (i.e., model building learning or model maintenance learning)
should be achieved so that the advice received can be judged to be useful by
knowledge seekers. That is, in the process o f e-mail-mediated knowledge transfer
across boundaries, the achievement o f intended learning mediates the relationship
between argument quality and perceived usefulness o f the knowledge received, after
controlling for other factors in the work environment, including: knowledge holder
credibility, and knowledge seekers’ expertise and involvement. Drawing from the
theory o f knowledge reuse, a good advice-seeking request sent through e-mail (i.e.,
with the features o f situational context, explication o f learning intent, and emotional
incentive) facilitates the achievement o f knowledge seekers’ intended learning. Also,
knowledge exploration boundaries moderate the relationship between the use of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
features o f advice-seeking requests and achievement o f intended learning. The
hypotheses that are posed are depicted in Figure 1.1
Figure 1.1 Conceptual Model
H2a H3a
Boundary-H5-
Argument H 1
Quality
Perceived U sefulness
o f K now ledge
R eceived
H4
Boundary
H3
-HJr H2
Control Variable
- Holder credibility
- Seekers’ expertise
x: Seekers’ involvem ei
Factual Context Explication of
MB intent
Explication of MM
intent
Brief Emotional
Incentive
Detailed Emotional
Incentive
Factual and
Interpretive Context
Achievement of Intended Model
Building Learning
Achievement of Intended Model
Maintenance Learning_________
1.3 Overview o f Research M ethodology
Data collection took place at Company A, an organization o f 2,000
employees. Company A is a global consulting firm promoting economic growth and
higher living standards in developing countries. The firm has its headquarters in
the US and more than 60 field offices worldwide focusing on different projects.
Projects are undertaken based on geographical regions that different departments
focus on (i.e., Africa, Asia, Europe and Eurasia, Latin America and Caribbean, and
Middle East). The research site was chosen because the nature o f Company A ’s work
requires its employees to seek cross-departmental knowledge, and they are heavily
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
reliant on communication technologies (i.e., e-mail distribution lists in this study) to
get their problems solved. A qualitative study was conducted as an initial step to
understand the impact o f contextual information in computer-mediated knowledge
seeking environments. The second phase o f the research, a quantitative study
including survey and coding, was used to test the hypotheses.
The preliminary qualitative data collection consisted o f on-site observation,
face-to-face interviews, and reviews o f e-mail examples from the participants.
Seventy-two people were interviewed including Project Administrators (PA),
Assistant Project Administrators (APA), Project Managers (PM), and Senior Vice
Presidents (SVP). They were asked about the experience o f seeking advice and
providing advice via e-mail. From these interviews, I learned that the nature o f the
employees’ work makes them extremely reliant on e-mail for getting their work
done. One major purpose o f these e-mails is to seek work-related advice (i.e.,
guidance, recommendations, and suggestions for how to solve non-administrative
problems). In order to better facilitate various projects all over the world, Company
A has recently built up their Practice Network system (e.g.,
Agriculture/Agribusiness; Democracy & Governance; Finance and Banking; etc)
across regions. Employees at Company A are encouraged to join the Practice
Network e-mail groups that they are interested in or specialized in. Knowledge
seekers at Company A reported often receiving e-mails from knowledge holders that
contained irrelevant information or suggestions. Also, informants reported that
knowledge sharing via e-mail among Company A employees was inefficient. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
problem w ith inadequate sharing o f advice appears to stem from the lack o f sharing
adequate contextual information. Respondents reported that e-mail requests that
included clearly specified contextual information seemed to attract more useful and
relevant advice from the knowledge holders.
Next, a quantitative study was conducted to further test the hypotheses and
the proposed model. The empirical study was conducted by distributing
questionnaires to advice seekers, and coding knowledge seekers’ e-mail messages for
context included when sending out the request for advice (i.e., coding e-mails
according to three dimensions o f advice-seeking definition in the e-mail solicitation
messages). Data were collected on argument quality, achievement o f intended
learning, knowledge usefulness, boundary information, and control variables using
the questionnaires. Data on various features o f advice-seeking requests were
obtained based on coding the knowledge seekers’ e-mail messages. This dissertation
study focuses on examining how to phrase e-mail solicitations to bridge the thought
world barriers and therefore facilitate achievement of intended learning. Given the
purpose o f this research study, only the e-mail messages sent by knowledge seekers
were coded. Access was provided for me to monitor the regional e-mail groups and
practice networks (i.e., e-mail groups formed by employees based on their
specialties) in which members sent out e-mail requests for advice. W hen a company
employee posted a request for help on one o f the practice networks (or regional
e-mail distribution lists), I asked them to complete a questionnaire. This
questionnaire asked the knowledge seekers what they intended to learn, and the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
extent o f their involvement and expertise (control variables). A response rate of
89.7% was obtained. One week later, Questionnaire II was sent to those who have
responded to Questionnaire I. The second questionnaire was intended to understand
what the knowledge seekers actually learned w ith the responses received, as well as
the perceived knowledge usefulness, argument quality, and the holder credibility.
Informants were asked to send back the actual e-mail inquiry and the most useful
e-mail advice (or, the latest e-mail response received if none o f the e-mail advice was
useful) as the basis for their responses. Each o f the e-mail messages was coded
according to the use o f contextual information (following the coding instructions)
and number o f boundaries. One hundred thirty-six employees at company A were
invited to complete the survey. O f these 136 employees, 117 responded to both
Questionnaire I and Questionnaire II, yielding a response rate o f 86%. Out o f these
117 responses, 8 were discarded and 109 were available for data analysis. Details are
discussed in Chapter 4.
1.4 Summary o f Findings
The goal o f this study is to test the mediation effect o f the achievement o f
intended learning on the relationship between argument question and perceived
usefulness o f the knowledge received in online communities. Another goal is to
examine the relationships between features o f advice-seeking requests in electronic
solicitations and achievement o f intended learning. Based on the objectives of this
dissertation and the sample size, Partial Least Squares (PLS) was used as the primary
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
data analysis technique (Chin, 2003). It was found that achievement o f intended
learning (i.e., model building learning, model maintenance learning) mediates the
relationship between argument quality and perceived knowledge usefulness
(Hypotheses 1). Further, the results confirmed that factual and interpretive context
and explication o f MM intent contribute to achievement o f intended model
maintenance learning (Hypotheses 2 and 3). Also, factual context and explication o f
MB intent are found to contribute to achievement o f intended model building
learning (Hypotheses 2a and 3a). Moreover, the relationship between situational
context and achievement o f intended learning is moderated by the number o f
boundaries that need to be crossed during the process o f electronic knowledge
transfer (Hypothesis 5). Boundary also moderates the relationship between
explication o f learning intent and achievement o f intended learning (Hypothesis 5).
The relationship between emotional incentive and achievement o f intended learning
was not supported by the results (Hypotheses 4 and 4a). The moderation effect of
boundary on the relationship between emotional incentive and achievement of
intended mental model learning was not supported, either (Hypothesis 5).
1.5 Summary o f Contributions
The findings o f this study provide several contributions to theory and future
research. The study also provides several implications for practice. This section
discusses the theoretical and practical implications, and avenues for future research.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
The study has several theoretical implications. First, the dissertation research
is an initial study addressing boundary barriers in knowledge transfer through
electronic distribution lists in online communities. Knowledge workers with similar
expertise or interests often belong to same electronic distribution lists (i.e., online
communities). Knowledge sharing facilitated by electronic distributions lists usually
induces different thought worlds that are distinguished by different e-mail
distribution lists. Current IS studies focus on e-mail-mediated (vs. electronic
distribution list) knowledge transfer (e.g., Sussman & Siegal, 2003) and develop an
information adoption theory to explain advice seeking in the e-mail settings by
emphasizing the effect o f argument quality and source credibility. However, this
research ignored thought world barriers that often create difficulties in
e-mail-enabled advice seeking processes. Drawing from the theory o f thought worlds
(Dougherty, 1992) and the theory o f knowledge reuse (Markus, 2001), this study
modifies the information adoption model by suggesting strategies to bridge thought
world barriers in cross-boundary knowledge access in online communities. Second,
the findings from this dissertation study also contribute to the thought world
literature. Despite the notion of interpretive barriers associated w ith knowledge
transfer across boundaries, Dougherty (1992) did not examine in depth how to
resolve the barriers. Literature on knowledge reuse (Markus, 2001) was drawn upon
to make suggestions - defining advice-seeking requests with critical features:
situational context, explication o f learning intent, and emotional incentive. The
findings from this study further suggest these features may facilitate knowledge
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
seekers achieving their intended learning, and then induce useful knowledge transfer.
Third, the findings from this dissertation study contribute to the discussion about the
value of technology in knowledge transfer. Technologies have been touted as a
means to widely disseminate diverse knowledge across organizational boundaries
(e.g. Alavi & Leidner, 2001; Huber, 1990). However, much o f the research to date
has shown that using technologies under such circumstances is problematic (Alavi &
Leidner, 2001; Markus, 2001). The findings from this dissertation study suggest that
proper employment o f strategies (i.e., use o f appropriate features o f advice-seeking
requests in electronic solicitations) could help bridge the barriers to cross-boundary
knowledge access supported by technology.
This dissertation study opens up new avenues for future research. For
example, this dissertation focuses on examining the impacts o f the choice o f learning
made (model building vs. model maintenance) in cross-boundary knowledge transfer
in online communities, but not specifically on predicting this choice. It w ill be
interesting to examine why knowledge workers would choose one type o f learning
over the other. Also, it would be fruitful to examine the views o f knowledge seekers
and holders simultaneously in a similar study to investigate how different dimensions
in advice-seeking requests affect knowledge holders’ perceptions differently. Future
research should also investigate in-depth the impact o f specifying emotional
incentive in technology-mediated knowledge transfer. In addition, the findings from
this study suggest that proper strategies employed in using technology (e.g., features
o f advice-seeking requests in electronic solicitations) may facilitate bridging barriers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
in cross-boundary knowledge transfer processes. Future research m ight investigate
how to incorporate these strategies into designing technologies for technologies to
facilitate cross-boundary knowledge access. In addition, findings indicated that most
employees agreed that the company highly encourages knowledge sharing within the
organization. It would be interesting future research to consider whether technology
features that support cross-departmental understanding can com pensate for less
supportive environments.
This study suggests practical implications as well. First, when knowledge
workers engage in technology-mediated knowledge transfer across boundaries,
thought world barriers make it hard for knowledge seekers to receive useful
knowledge in online communities. Current theories on advice-seeking over an
organizational computer network ignore whether the advice comes from the same or
a different thought world (e.g., Sussman & Siegal, 2003), thus assuming that advice
across thought worlds is the same as seeking advice from the same thought world. In
online communities, knowledge seekers m ay or may not know who might read the
electronic queries given the solicitations are sent to a large electronic distribution list.
Knowledge seekers should not assume knowledge holders always interpret their
questions from the same thought world. In order to achieve their learning objectives
(i.e., intended learning) so as to perceive the knowledge received as useful, they must
communicate and convey the objectives and/or expectations to knowledge holders
since boundaries create barriers for holders to identify learning needs. Also, it is very
important for knowledge seekers to deploy strategies (specification o f situational
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
context and explication o f learning intent) to facilitate the very early stage o f
electronic knowledge transfer - electronic knowledge seeking, especially when there
are boundaries to be crossing.
The results o f this dissertation study provide a simple and feasible way for
employees to better share knowledge through electronic distribution lists and
increase their effectiveness and efficiency in electronic knowledge sharing. The
results o f this dissertation study should lead to feasible methods for more effective
e-mail usage in organizations, and a better practical understanding o f knowledge
sharing through electronic distribution systems. For instance, in order to achieve
model building learning (in which m ental models are changed to accommodate new
information), employees who send out electronic inquiries across boundaries should
include factual situational context information and high explication o f learning
intent. In contrast, in order to achieve model maintenance learning (in which new
information fits into existing mental models and confirms them), employees should
provide interpretive situational context information and low explication o f learning
intent in their e-mail solicitations crossing boundaries. These guidelines could be
included in a Best Practices document or training program for Practice Network
facilitators and Company A employees. The company would also benefit from a
better use o f electronic distribution lists for fostering its knowledge-sharing culture
in CMC.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
1.6 Roadmap to the Dissertation
Chapter 2 introduces the dissertation study by providing a statement o f the
problem and discussing background literature from thought world, knowledge reuse,
cooperative learning (from psychology), and IS research. Chapter 3 uses the theory
o f thought worlds and the theory o f knowledge reuse to describe the conceptual
model and hypotheses. The research methodology is described in Chapter 4, and
includes a detailed discussion o f the qualitative (i.e., interview, observation, review
o f e-mail examples) data collection methods and quantitative (i.e., survey, and
coding) data collection methodology. The results are presented in Chapter 5. Chapter
6 discusses the results, and details the limitations and the contributions to theory,
future research, and practice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
CHAPTER 2 INTRODUCTION
In the information age, organizations are under more pressure than before to
sustain a competitive advantage in their respective industries. A m ajor source o f
competitive advantage for most organizations comes from knowledge assets (Alavi
& Leidner, 2001; Grant, 1996). W ith the evolution o f knowledge management, we
have seen increased emphasis placed on understanding and supporting processes o f
knowledge creation and transfer within and across groups in organizations (Cross,
Rice, & Parker, 2001). Different groups (i.e., communities) o f knowledge workers
create knowledge exploration boundaries for knowledge transfer (Bechky, 2003;
Cross, Rice, & Parker, 2001; Rosenkopf & Nerkar, 2001). Organizations face
continued difficulties when employees seek, share, transfer, or otherwise reuse each
other’s knowledge across boundaries which are created by different departments or
specialties (Argote & Ingram, 2000; Szulanski, 1996). By knowledge, I adopt a
broad definition consistent with prior research (Beccerra-Femandez & Sabherwal,
2001; Grant, 1996) to include explicit knowledge such as drawings, analytic results,
and scientific journals, as well as tacit knowledge such as insights, intuition, and
implied assumptions (Kogut & Zander, 1992; Polanyi, 1967).
W ith the advent o f communication technology such as electronic bulletin
boards, instant messaging, and hypertext browsers (e.g., the W orld W ide Web),
knowledge workers can now easily communicate electronically w ith other
individuals in distributed networks (Kettinger & Grover, 1997). Further,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
organizations have been increasingly providing technological support for knowledge
sharing, m ost commonly via Intranets (Hollingshead et al., 2002). In 1996, two thirds
of Fortune 500 companies had intranets in operation (Head, 2000).
According to Hollingshead et al. (2002), Intranets refers to “company
websites designed for internal use, are an important technological innovation in
many organizations that can aid in knowledge management, expertise recognition
and communication” (p. 335). The m ost widely used function o f Intranet is
electronic distribution list (Levitt & Mahowald, 2001). Electronic distribution list
enables individuals to post electronically a message to a large num ber o f individuals
in online communities. “As organizations become less hierarchical, the need for
non-impositional forms o f knowledge transfer, such as advice giving and receiving,
is likely to increase” (Sussman & Siegal, 2003, p. 49). In m ost ways, knowledge
transfer is more effective when it occurs in a face-to-face context rather than a
mediated one (Brown & Duguid, 1991; Lave & Wenger, 1991; Nadler et al., 2003;
Zack & McKenny, 1995). Despite the increasing attention from scholars, it is not
clear why electronic settings often lead to unsuccessful knowledge transfer and how
knowledge seekers can achieve their learning needs crossing boundaries. Thus, this
dissertation research focuses on a particular type o f e-mail: Electronic solicitation
through which knowledge workers seek advice through electronic distribution lists
(i.e., the responding e-mail is expected to contain desired advice, suggestions, and
recommendations when transferred to knowledge seekers) from others. Advice is
generally viewed as “information that communicates an opinion about what could or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
should be done about a problem or issue” (Sussman & Siegal, 2003, p.49). In this
study, any opinion or recommendation that knowledge seekers receive in response to
their queries would be construed as advice. More specifically, I limit the discussion
to advice seeking through electronic distribution lists across intra-organizational
boundaries.
Current theories o f electronic knowledge transfer ignore the boundary
barriers by emphasizing the role o f argument quality in determining a knowledge
seeker’s perceived usefulness o f the knowledge received in CMC. Instead, the
theories assume knowledge holders and seekers share the same thought worlds
(Dougherty, 1992) and thus knowledge holders understand seekers’ learning needs.
In addition, it is not clear how to bridge the barriers (created by different thought
worlds) to cross-boundary knowledge access in online communities. Bridging
thought world barriers is essential to understand how to phrase advice-seeking
requests in electronic solicitations to receive useful knowledge when there are
different needs (or, objectives) in the process o f knowledge seeking (Markus, 2001).
Thus, this area o f research is o f particular interest in this dissertation study.
This chapter begins by describing a practical problem in an organizational
setting, as the motivation for the study (section 2.1). A review o f the current IS
literature highlights the relationship between argument quality and perceived
usefulness o f the knowledge received through e-mail (section 2.2). The gap in the
current literature about electronic knowledge transfer is then discussed drawing from
the theory thought worlds (section 2.3). In order to bridge the thought world barriers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
in electronic knowledge transfer processes, the importance o f meeting knowledge
seekers’ intentions is examined (section 2.3). This is followed by a brief introduction
to the solutions o f how to facilitate the knowledge seekers’ search intentions to be
met - defining the advice-seeking request in electronic solicitations (section 2.4).
Coupling the theory o f thought worlds and M arkus’ theory o f knowledge reuse,
detailed discussion about achieving knowledge seekers’ search intentions and
defining the advice-seeking request w ill be provided in Chapter 3. The chapter ends
w ith a summary (section 2.5).
2.1 M otivation for Study
Tom is a Project Assistant in a US government contracted international
consulting firm. Before his business trip to Guyana, he decided to send out a
broadcasting e-mail to the corporate headquarter office (i.e., to several e-mail
distribution lists), o f which many people have international experiences. In the
message, he asked the question: “I’m going to visit Guyana for two weeks. Could
you please provide any information?” A senior person in the headquarter office,
“Jane”, was willing to help but was confused by the email, and responded with a
flurry of questions: “What are you going to do there?” “W hat do you want to know?”
“We have three reports related to Guyana. Two huge ones are very important. Do
you need that kind o f information?” She needed a few e-mail exchanges with Tom
and numerous man-hours before it became clear to her how she could help. A few
days were wasted in the process. It is obvious that Tom did not articulate what he
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
intended to learn and the context in which he planned to use the knowledge he
sought.
This scenario exemplifies one problem of electronic knowledge transfer in
distributed networks. “In geographically dispersed organizations, employees cannot
always get useful advice from their local colleagues” (Constant et al., 1999, p. 415).
Thus, increasingly, employees access cross-boundary knowledge electronically in
online communities. Defining the advice-seeking request is critical in electronic
knowledge transfer crossing boundaries (Markus, 2001). W ithout a properly defined
question in CMC across boundaries, there can be a number o f unpleasant
consequences (e.g., time delay, information overload, useless knowledge) (e.g.,
Wastell, 1999). In particular, knowledge workers may need to spend extra time and
effort to extract underlying assumptions before knowledge seekers and holders reach
the same horizon o f understanding. Knowledge holders might stop contributing
owing to the unanticipated time and effort that will be involved.
In the information age, knowledge work is becoming increasingly dependent
on electronic knowledge transfer (Sussman & Siegal, 2003). Electronic knowledge
work has been attracting increasing attention (e.g., Dennis & Kinney, 1998; Markus,
1994; Sussman & Siegal, 2003). Knowledge transfer through electronic distribution
lists increases the need for advice to be sought across boundaries (Kraut et al., 2001).
Nonetheless, knowledge access crossing boundaries is difficult due to interpretive
barriers (Dougherty, 1992). Current literatures on electronic advice-seeking ignore
whether the advice comes from the same or a different thought world (e.g., Sussman
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
& Siegal, 2003). Instead, they assume that advice received across thought worlds is
fundamentally the same as seeking advice from the same thought world. Literature
on thought worlds indicates different cognitive processes are involved to seek advice
across thought worlds (Dougherty, 1992). Thus, the fundamental assumption of the
electronic advice-seeking literature is questionable.
Cross-boundary knowledge seeking requires access to insights from a variety
o f knowledge specialties (i.e., online communities) (Rosenkopf & Nerkar, 2001).
Dougherty (1992) calls these knowledge specialties thought worlds. According to
Dougherty, knowledge workers across boundaries not only know different things
(i.e., specialty; Dougherty called it “funds o f knowledge”), but they also know things
differently (i.e., interpretation, Dougherty called it “system o f meaning”). As such,
different thought worlds create barriers for knowledge holders to understand
knowledge seekers’ needs across boundaries. Meeting knowledge seekers’ needs is
critical in successful knowledge transfer (Markus, 2001). Given the difficulties
created by thought world differences in cross-boundary knowledge transfer, there is
a need for research to recommend strategies in bridging thought world barriers in
CMC to meet different knowledge-seeking needs. The key is to define the electronic
advice-seeking request (Markus, 2001).
W ithout a well-defined advice-seeking request, the received knowledge
through e-mail m ay be irrelevant or relatively useless, even though knowledge
holders m ay have taken time to follow up with elaborated responses and high quality
arguments. Returning to the Tom & Jane’s example, Tom stated to me, “I received
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
many responses through e-mail, m ost o f which might have been useful in some
circumstances other than those that prom pted me to send the e-mail” (paraphrased).
It is obvious that Tom did not receive advice that matched his seeking objectives and
therefore, he cannot achieve what he intended to leam. Also, an inappropriate
question in the electronic solicitation created the problem o f information overload.
Although not knowing Tom ’s statement, Jane in the example above coincidently
stated to me: “W e could definitely help to make sure that useful and relevant
information is being acquired through e-mail if we can understand his situation and
what he needed”. Proper and precise advice-request in e-mail solicitations will help
knowledge holders understand knowledge seekers’ needs so that holders can better
help seekers (Augier et al., 2001; Cossette, 1998). In short, e-mail-supported
knowledge seeking without appropriate advice-seeking requests results in
unnecessary effort from knowledge workers and inefficient knowledge sharing.
Moreover, knowledge holders may resist or be reluctant to contribute due to
unanticipated costs o f making good advice.
The problem is even more salient in knowledge transfer processes facilitated
by electronic distribution lists. Knowledge is considered to be embedded in different
repositories (Argote & Ingram, 2000) (i.e., different online communities) and created
in different thought worlds (Dougherty, 1992). E-mail-mediated knowledge transfer
is based on asynchronous medium with fewer social context cues and weakened
social presence o f communicators compared with the face-to-face interactions. In
online communities, knowledge seekers may or may not know potential knowledge
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
holders involved in electronic distribution lists. As such, e-mail m akes it difficult to
transmit equivocal messages and brings limitations to mitigate thought world
differences between knowledge seekers and holders (Yamauchi et al., 2000).
Consequently, it is essential to define the advice-seeking request properly so that
knowledge holders might have resources (i.e., directions) to produce high quality
knowledge to meet knowledge seekers’ needs (Markus, 2001). W hen knowledge
transfer requires crossing more intra-organizational boundaries, more radical changes
in meaning may occur, and therefore, it is more important to define the
advice-seeking request appropriately. Hence, in this dissertation study, I limit the
discussion to electronic advice seeking across intra-organizational boundaries in
online communities, a much understudied area o f CMC. W ithin the proposed
conceptual model, advice-seeking is conceptualized as the pursuit o f desired advice
or information about a target problem.
2.2 Electronic Knowledge Transfer
E-mail is a form of asynchronous CMC that has become commoditized in
knowledge-based organizations (Sussman & Siegal, 2003). E-mail “uses computer
text processing and communication tools to provide a high speed information
exchange service” (Sproull & Kiesler, 1986, p. 1493). M ore than 10 billion e-mail
messages are now exchanged daily, and this quantity is expected to increase to 35
billion by 2005 (Levitt & Mahowald, 2001). In a study o f managerial e-mail use,
27% of managers reported using e-mail for equivocal communication such as
exchanging subjective views, and 14% use e-mail to influence or sell an idea
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
(Markus, 1994). This evidence suggests that knowledge transfer via e-mail is a
rapidly emerging phenomenon (Lee, 1994; M arkus, 1994; Sussman & Siegal, 2003)
and goes beyond simple information exchange.
The growth o f CMC and the Intranet, has brought with them the need for
unique mechanisms for seeking information from others (Ramirez et. al., 2002) -
e-mail distribution list. As the most pervasive form o f CMC, organizational use o f
e-mail has been w idely studied for its relationship to media richness (Dennis &
Kinney, 1998; Fulk, 1993; Lee, 1994; Markus, 1994; Rice, 1992; Zack, 1993). These
theoretical approaches failed to take into account “the active role o f individuals in
the communication process, much less information seeking, opting instead to
emphasize channel effects” (Ramirez et al., 2002, p. 215). Although most CMC
environments eliminate or severely reduce nonverbal and contextual information
available to address uncertainty, outline objectives, develop relationships, and bridge
barriers, such environments offer alternative mechanisms for acquiring useful advice
from others (Ramirez et. al., 2002).
In a recent study, Sussman and Siegal (2003) examine individuals’
perceptions o f usefulness o f the knowledge received through e-mail and how
individuals incorporate received advice into their actions. The authors found that
knowledge seekers are more likely to perceive the advice (received from knowledge
holders) as useful when the knowledge holder does not just present the information
but also offers a cogent argument for its worth (what has been called “argument
quality”). In their study, argument quality refers to the completeness, accuracy, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
consistency o f the content o f the responding e-mail, which contains advice. Source
credibility refers to “a message recipient’s [i.e., knowledge seeker] perception o f the
credibility o f a message source, reflecting nothing o f the message itse lf’ (Sussman &
Siegal, 2003, p. 51). The authors suggest that argument quality and source credibility
represent different cues in judging the usefulness o f the knowledge received through
e-mail.
According to Sussman and Siegal’s (2003) information adoption theory,
central and peripheral cues1 influence perceived knowledge usefulness (see Figure
2.1 for Sussman and Siegal’s model predicting knowledge usefulness). This was
drawn upon from the Elaboration Likelihood Model. The ELM developed by Petty
and Cacioppo (1986) states that “in different situations, different message recipients
will vary in the extent to which they cognitively elaborate on a particular message,
and these variations in elaboration likelihood affect the success o f an influence
attempt” (Sussman & Siegal 2003, p. 50). Specifically, high levels o f elaboration
represent a central route to influence, while low levels result in a peripheral route.
According to the ELM, when a knowledge worker is “able and willing to cognitively
elaborate on a persuasive communication” (i.e., the situation o f high-elaboration
likelihood), argument quality (contained within the communication) is the critical
determinant of perceived usefulness o f the knowledge received. Under the conditions
o f low-elaboration likelihood (i.e., a knowledge seeker is either unable or unwilling
to process the arguments received), peripheral cues are critical in predicting the
1 Peripheral cues will not be the interest of this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
perceived usefulness o f the knowledge received. Among different peripheral cues
that have been studied (e.g., source’s attractiveness and likeability), source
credibility is the one that may be relevant both within the context o f CMC and within
more general knowledge work (Sussman & Siegal, 2003). W ith a field study in a
consulting firm, the authors found that argument quality is the critical central cue for
knowledge seekers to perceive the received knowledge through e-mail as useful
(Sussman & Siegal, 2003).
Sussman and Siegal’s model presents an initial model predicting the
antecedents o f informational usefulness in e-mail-facilitated advice-seeking. Current
knowledge transfer literature suggests other critical factors are ignored by Sussman
and Siegal to explain the process o f cross-boundary knowledge transfer in the e-mail
settings. One critical factor ignored involves the thought w orld barriers (created by
the boundaries that need to be crossing) in the e-mail-mediated advice-seeking
processes (Dougherty, 1992). When thought worlds are being crossed, such factors
as a seeker’s intended learning become more important since this will not necessarily
be understood by the knowledge holder. W hile quality o f an argument may be an
important factor affecting knowledge usefulness, if the argument quality is not
relevant to a knowledge seeker’s intended learning, argument quality will have
minimal useful impact. This dissertation begins with Sussman and Siegal’s (2003)
model and then discusses the gap in the current e-mail-facilitated advice-seeking
literature. M arkus’ theory o f knowledge reuse will then be discussed with respect to
how to bridge the barriers by defining advice-seeking requests in e-mil solicitations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
In sum, the theory o f thought worlds coupled w ith the theory o f knowledge reuse
w ill be employed to enrich the information adoption theory in e-mail-mediated
advice-seeking crossing boundaries.
Figure 2.1 Sussman and SiegaFs Model: E-mail-mediated Advice-seeking
Recipient Involvement
Recipient Expertise
Argument Quality
Source Credibility
Knowledge
Usefulness
Sussman & Siegal (2003)
Information Systems Research
2.3 Knowledge Transfer across Boundaries
Knowledge transfer within organizations can be defined as the process
through which one organizational unit (e.g., person, group, department, or divisions)
is affected by the experience and knowledge o f another (Argote, 1999; Argote &
Ingram 2000; Szulanski, 1996). A process o f cross-boundary knowledge transfer
requires access to insights from a variety o f knowledge specialties. Dougherty (1992)
calls these knowledge specialties thought worlds.
Knowledge holders and seekers need to transcend their local thought worlds
when they are from different departments and/or hold different specialties
(Dougherty, 1992). In the electronic settings, knowledge seekers m ay or may not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
know the potential knowledge holder and cannot perceive whether knowledge
holders understand their words or question. Knowledge holders m ay view knowledge
seekers “central issues as esoteric, if not meaningless” (Dougherty 1992 p. 182).
Thus, even if knowledge holders put a major effort into producing high quality
argument, it might not be useful for knowledge seekers as the argument does not
match seekers’ learning intentions. In building a theory o f knowledge reuse, Markus
(2001) also emphasizes the significance o f matching knowledge seeker’s needs in
successful knowledge transfer processes.
Successful knowledge transfer requires knowledge holders to produce
knowledge that meet knowledge seekers’ learning needs (Markus, 2001). Given
different expectations, knowledge seekers m ay conduct different search behavior or
need different facilitation activities (e.g., abstracting, authoring, indexing, etc.)
(Markus, 2001). They may learn differently from these different search behaviors.
For knowledge transfer to occur, learning must happen in the mind o f the recipient
(Sussman & Siegal, 2003). The cognitive school o f learning supplements the theory
o f knowledge reuse to detail different search intentions. Cognitive theorists make
similar statements about the relationship between search behavior and learning
needs. Knowledge seekers without a particular problem to solve or question to
answer may browse information (i.e., scanning) to achieve their learning needs
(Vandenbosch & Higgins, 1996; Webb & Palinscar, 1996). In comparison,
knowledge seekers in need o f advice to solve a particular problem may conduct
“focused search” (Huber, 1991; Webb & Palinscar, 1996) to reach their learning
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
objectives. If the knowledge received cannot help knowledge seekers to achieve
what they intend to learn, anxiety may occur, which “alerts the organism to the
presence of a significant mismatch between its desired goals and the current
situation” (W astell, 1999, p. 583). As such, the knowledge seeker m ay be reluctant to
process and accept the knowledge received, and engage in “defensive, antileaming
processes” (W asted, 1999). As a result, the knowledge received through e-mail is
unlikely to be perceived as useful by knowledge seekers. The challenge then
becomes how to facilitate the achievement o f knowledge seekers’ intended learning
so as to induce useful knowledge received through e-mail.
2.4 How to Facilitate Achievement o f Intended Learning by Bridging Boundary
Barriers
Knowledge transfer crossing boundaries entails thought world barriers which
make it difficult for knowledge holders to understand knowledge seekers’ needs
(Dougherty, 1992). M arkus’ (2001) theory o f knowledge reuse suggests that defining
the advice-seeking request is the key. Knowledge seekers across boundaries are often
unable to define the advice-seeking request properly (Markus, 2001). This may
create increased difficulty for knowledge holders to understand the seekers’ search
objectives.
Defining the advice-seeking request properly is essential for knowledge
seekers to receive knowledge that matches their learning needs (Markus, 2001).
According to M arkus (2001), “one characteristic separating experts from novices is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
that experts know what questions to ask” (p. 61). Drawing from M arkus’ (1992)
theory o f knowledge reuse, features o f w ell-defined advice-seeking requests include:
(1) situational context; (2) explication o f learning intent; and (3) emotional incentive
in the e-mail solicitations. Situational context helps knowledge holders understand
assumptions and background information surrounding the question. It has long been
noted that situational context is important in successful knowledge transfer (Markus,
2001; Nonaka et al., 2000; Taylor, 1996; T e’eni, 2001). Explication o f learning
intent identifies knowledge seekers’ intentions if they are open to alternate
perspectives, which is associated with different learning needs. Explication of
learning intent helps knowledge holders understand knowledge seekers’ learning
intentions and search objectives; and therefore, they can provide useful as well as
high quality advice. Emotional incentive refers to the provision o f emotion or
attitude information so as to motivate knowledge holders to respond and contribute.
Lack o f appropriate incentives to contribute was implicated in the knowledge
transfer failure (Szulanski, 1996). The three factors will be discussed in more detail
in Chapter 3.
2.5 Summary
In sum, achievement o f intended learning is essential for obtaining useful
knowledge from prior research. In Sussman and Siegal’s (2003) study, it is suggested
argument quality leads to perceived usefulness o f the knowledge received through
e-mail. Nonetheless, a brief review o f the theory o f thought worlds and M arkus’
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
theory o f knowledge reuse suggests that we should consider thought world barriers
in e-mail-mediated advice-seeking crossing boundaries, and further incorporate
achievement o f intended learning into Sussman and Siegal’s (2003) model to present
a more comprehensive framework o f understanding cross-boundary knowledge
transfer in the CMC context. In addition, strategies o f defining the advice-seeking
request should be employed to facilitate achievement o f intended learning.
The purpose o f this dissertation research therefore is to present a conceptual
model and empirically investigate cross-boundary knowledge access bridging
thought world barriers in the e-mail settings. Specifically, in this study, I attempt to
answer this research question:
During electronic solicitations by knowledge seekers, how should the advice-seeking
request be defined when organizational boundaries are crossed fo r the seeker to
achieve his/her intended learning and obtain knowledge he or she judges useful?
The next chapter explains the literature in more detail. First, it presents the
theory o f thought worlds and how it addresses the gap in current theories o f
e-mail-mediated knowledge transfer. The theory of knowledge reuse is then
discussed to highlight the importance o f achieving intended learning. Different
dimensions o f a well-defined advice-seeking request are also discussed drawing
upon M arkus’ theory o f knowledge reuse. It is followed by a review o f the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
conceptual model and the hypotheses. The chapter ends w ith a summary o f the
proposed model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
CHAPTER 3 LITERATURE RESEARCH AND CONCEPTUAL
MODEL
This chapter opens w ith a presentation o f the theory o f thought worlds. An
explanation o f the barriers individuals encounter when accessing knowledge from
different thought worlds is outlined (section 3.1). It is followed by the elaboration
o f M arkus’ theory o f knowledge reuse (section 3.2) which provides insights to bridge
the thought world barriers in electronic knowledge transfer processes. This includes
a description o f knowledge reuse roles, knowledge reuse processes, and different
types of knowledge reuse situations. Also, how to apply the theory o f knowledge
reuse is generally discussed. Further elaboration o f the theory suggests that meeting
knowledge seekers’ needs is critical in successful e-mail-mediated knowledge
transfer processes (section 3.3). Learning literature supplements the theory o f
knowledge reuse to explain the importance o f matching knowledge seekers’ search
intentions. Using both theories, it is hypothesized that achievement o f intended
learning mediates the relationship between argument quality and perceived
usefulness o f the knowledge received through e-mail. Also, it is proposed that the
w ay advice-seeking requests are phrased in e-mail solicitations to cross thought
world barriers contributes to the achievement o f intended learning (section 3.4).
Three dimensions o f advice-seeking requests are suggested, drawn from M arkus’
theory o f knowledge reuse: situational context, explication o f learning intent, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
emotional incentive. Further, the more knowledge transfer boundaries to be crossing,
the more important it is to specify the three elements to define the advice-seeking
request. Control variables are then discussed (section 3.5). This is followed by the
description o f the complete conceptual model (section 3.6). The chapter ends with a
summary (section 3.7).
3.1 Theory o f Thought W orlds
A process o f cross-boundary knowledge transfer requires access to insights
from a variety o f knowledge specialties - referred to as thought worlds in Dougherty
(1992). “A thought world is a community o f 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 w ithin an organization that are
divided along departmental boundaries, such as manufacturing, marketing, sales, and
technical departments. He found that 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. Prior research on e-mail-mediated knowledge transfer
ignored the thought world barriers created by assuming knowledge holders and
seekers share the same thought worlds and thus understand knowledge seekers’
learning needs.
Scholars have investigated some important factors influencing knowledge
transfer using CMC, such as characteristics o f knowledge seekers (Stamm & Dube,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
1994; Sussman & Siegal, 2003); characteristics o f knowledge holders (Faraj &
Sproull, 2000; Sussman & Siegal, 2003; W u & Shaffer, 1987); characteristics o f the
communication medium (Lee, 1991; Markus, 1994); and properties o f knowledge
(e.g., Nonaka et. al., 2000). Recently, Sussman and Siegal (2003) propose a model
predicting the antecedents o f knowledge usefulness in e-mail-facilitated
knowledge-seeking processes. The authors found that argument quality o f the advice
received through e-mail is crucial for knowledge seekers to judge the advice
(received from knowledge holders) as useful. This research did not capture one o f
the unique features o f e-mail-facilitated knowledge transfer processes - knowledge
holders and seekers’ different thought worlds. Knowledge transfer in the e-mail
settings increases the need o f knowledge access across boundaries (Kraut et al.,
2001). Nonetheless, knowledge access crossing boundaries is difficult due to thought
world barriers (Dougherty, 1992). Current literature on e-mail-facilitated
advice-seeking ignores whether the advice comes from the same or a different
thought world (e.g., Sussman & Siegal, 2003). Instead, it is assumed in the current
literature on e-mail-facilitated advice seeking that advice across thought worlds is the
same as seeking advice from the same thought world. However, literature on thought
worlds indicates different cognitive processes are involved when seeking advice
across thought worlds (Dougherty, 1992). Thus, the fundamental assumption
underlying current theories o f e-mail-facilitated advice seeking should be modified
to provide a more adequate model predicting knowledge usefulness in the e-mail
settings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
According to the theory o f thought worlds, the barriers o f knowledge
transfer result from the different a) funds o f knowledge, b) systems o f meaning, and
c) readiness for directed perception that comprise each thought world. Knowledge
workers in different departments and/or with different specialties hold different
thought worlds. Each thought world represents a different fund o f knowledge (i.e.
knowledge workers know different things). A fund o f knowledge represents an
interest or area o f expertise that is rooted in the skills and experiences members o f a
thought world have developed over time (Dougherty, 1992). In addition, each of
the thought worlds also has a different system o f meaning (i.e. knowledge workers
interpret things differently). This presents another barrier, in that people belonging
to different thought worlds often interpret the same piece o f knowledge differently
(Dougherty, 1992). Moreover, each thought world also develops a readiness for
directed perception - an intrinsic harmony, where members o f a thought world
develop a particular point o f view based on the expertise they and other members o f
their thought world have developed over time (Dougherty 1992). Knowledge that is
acquired from different thought worlds operates under different systems o f meaning
may be rejected deliberately, even though it m ay have been useful knowledge.
Given all these barriers, it is difficult for knowledge seekers to access
cross-boundary knowledge. It is even harder for knowledge holders to produce
knowledge that meets knowledge seekers’ needs or provide knowledge that will be
perceived as useful by knowledge seekers. This suggests that Sussman and Siegal’s
(2003) model is not adequate to explain e-mail-supported knowledge transfer across
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
boundaries. Dougherty (1992) details differences among the thought worlds which
keep knowledge workers from synthesizing their expertise. Nonetheless, the theory
o f thought worlds does not explain in depth how to address thought worlds barriers
to improve knowledge w orkers’ performance to ensure successful knowledge
transfer. M arkus’ (2001) theory o f knowledge reuse provides insights to answer this
question.
3.2 Theory o f Knowledge Reuse
To consider strategies that facilitate electronic access to cross-boundary
knowledge (that is associated with thought worlds barriers), this dissertation study
draws from M arkus’ theory o f knowledge reuse. M uch o f the research that
conceptualizes knowledge transfer challenges has emphasized general processes that
organizations use to codify and transfer knowledge across boundaries (Bechky,
2003). Theory o f knowledge reuse makes specific suggestions how knowledge
seekers m ay facilitate knowledge holders to produce useful knowledge for future
reuse. Examining both successful and unsuccessful situations o f knowledge reuse,
Markus (2001) develops a theory o f knowledge reuse including: knowledge reuse
situations, knowledge reuse process, and the major roles in the knowledge reuse
process.
According to Markus (2001), three major roles are involved in the knowledge
reuse process: knowledge producer (who originate and document knowledge),
knowledge intermediary (who prepares knowledge for reuse and performs diverse
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
roles in dissemination and facilitation), and knowledge consumer (who retrieves the
knowledge and applies it in their own work). IS research has examine the role o f
intermediaries in the context o f electronic communication systems (e.g., Orlikowski
et al., 1995) and electronic meeting systems (e.g., Griffith et al., 1998). However, the
role o f intermediaries has not attracted enough attention (Markus, 2001).
According to the theory o f knowledge reuse, the knowledge reuse process
consists o f four phases: capturing knowledge, packaging knowledge, distributing
knowledge, and reusing knowledge (Markus, 2001). Further, the reusing knowledge
phase can be described in terms o f the following four activities: defining the search
question; search for, and location of, experts or expertise; selection o f an appropriate
expert or o f expert advice from the results o f the search; and applying the knowledge
(Markus, 2001). Among these four activities, defining the search question is the first
step and is critical for successful reuse. Similarly, it is essential for knowledge
seekers to conduct search when they send out e-mail inquiries. A more detailed
discussion o f defining the search question will be provided in the following section.
By reviewing both academic and practical writings on knowledge reuse,
Markus (2001) suggests four different types o f knowledge reuse situations
characterized by different knowledge reusers and the purposes o f reuse. The four
situations are: reuse by shared work producers (i.e., knowledge workers “produce the
knowledge themselves while working on a shared work product” p. 63), reuse by
shared work practitioners (i.e., knowledge workers sharing a community o f practice
produce knowledge for each other to use), reuse by expertise-seeking novices (i.e.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
knowledge workers access to experts and expertise), and reuse by secondary
knowledge miners (i.e., knowledge workers who “extract knowledge from records
that were collected by others” p. 71) (Markus, 2001). Among these different types of
knowledge reuse, expertise-seeking novices are similar to the knowledge seekers in
this study, as this situation often involves knowledge transfer across organizational
boundaries. This type o f knowledge workers m ay have great difficulty in defining
the advice-seeking request and locating suitable experts (Markus, 2001). The more
boundaries to be crossing, the more difficult it is for them to define the
advice-seeking request (Markus, 2001). Thus, a well-defined advice-seeking request
is especially important for this situation and this type o f reuser (i.e., knowledge
seeker in this dissertation study).
Examining the failures o f knowledge reuse, M arkus found that each type o f
knowledge reuser has different requirements for the existing knowledge repositories.
In addition, reusers ’ requirements often remain unmet given how repositories are
created. M arkus’ (2001) theory suggests that meeting knowledge seekers’
requirements or needs is critical for successful knowledge transfer processes.
Meanwhile, it requires a major effort from knowledge holders to repurpose the
knowledge repositories for the knowledge to be useful for new reusers (Markus,
2001). However, knowledge producers rarely have the resources and incentives to
repurpose the existing knowledge (Markus, 2001). Consequently, Markus (2001)
suggests some solutions for facilitate meeting knowledge reusers’ needs - defining
the search question and careful use o f incentives. Another solution suggested is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
appreciation o f intermediaries who “prepares knowledge for reuse by eliciting it,
indexing it, summarizing it, sanitizing it, packaging it” (Markus 2001, p. 61). This
solution has been integrated in defining the advice-seeking request as knowledge
seekers facilitate knowledge transfer by directing knowledge producers to targeted
advice. Under this particular circumstance, the knowledge seekers actually serve as
the intermediaries and indirectly help knowledge holders index and package the
knowledge. These solutions further suggest the features o f advice-seeking request
that facilitate transfer o f useful knowledge: (1) richness o f situational context, (2)
explication o f learning intent, and (3) emotional incentive.
In the following paragraphs, the theory o f knowledge reuse is drawn upon to
discuss these three dimensions in detail. The potential benefit each o f these
dimensions brings to meeting knowledge seekers’ needs when accessing knowledge
from different thought worlds is also discussed. Supporting evidence found in the
related literature adds to the discussion.
3.3 Importance of M eeting Learning Intentions
Thought world barriers suggest that Sussman and Siegal’s (2003) model is
not adequate to predict knowledge usefulness in the process o f electronic knowledge
transfer crossing boundaries in online communities. In this section, drawing from the
theory o f knowledge reuse and cooperative learning, it is explained why it is not
sufficient in a cross-boundary context for argument quality to lead directly to
knowledge usefulness. In addition, it is discussed why knowledge seekers’ intentions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
should be met and how the intentions may be achieved in the electronic settings to
predict knowledge usefulness. Learning literature is briefly reviewed to identify
different learning intentions.
3.3.1 Achieving Intended Learning in a Cross-boundary Context
Successful knowledge transfer requires that knowledge seekers’ needs are
met (Markus, 2001). Knowledge workers who seek advice are unlikely to have the
same objectives (or, intentions) for that inquiry. More specifically, knowledge
seekers may expect different benefits (e.g., solutions, meta-knowledge - pointers to
databases or people, validation o f plans or solutions, problem reformulation, etc)
(Cross, Rice, & Parker, 2001) or have different desired goals (Te’eni, 2001; Wastell,
1999) when seeking advice from other people. If knowledge seekers receive
knowledge that cannot match their search intentions, they m ay “drown in
unnecessary, unhelpful, or conflicting data” (Markus, 2001). In the following
paragraphs, literature on cooperative learning is coupled w ith the theory o f
knowledge reuse to explain why meeting knowledge seekers’ learning intentions is
essential for knowledge received to be judged useful by the seekers.
In order to facilitate different search intentions to be met, knowledge
seekers may conduct different search behavior, such as key word-indexing or
document browsing (Markus, 2001). They may in turn learn differently from these
different search behaviors. For knowledge transfer to occur, learning must happen in
the mind o f the recipient (Sussman & Siegal, 2003). Cognitive theorists make similar
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
statements about the relationship between search behavior and learning needs. From
a cognitive learning perspective, Vandenbosch and Higgins (1996) provide one o f
the few research discussions that offer considerable detail as to how different
learning intentions are associated with different search behaviors. Knowledge
seekers without a particular problem to solve or question to answer may browse
information (i.e., scanning) to achieve their learning needs (Vandenbosch & Higgins,
1996; Webb & Palinscar, 1996). In comparison, knowledge seekers looking for
specific information to solve a problem m ay need to conduct focused search (Huber,
1991; Webb & Palinscar, 1996) to reach their learning objectives.
The cognitive school o f learning supplements the theory o f knowledge reuse
to detail different search intentions. Individual level intended learning can be
achieved when “solutions from one unit are matched to problems o f an individual
from another unit (problem-solution exchange)” (Goodman & Darr, 1998, p. 419).
This suggests that desired learning occurs when the received solutions are targeted to
intended problem solving - meet knowledge seekers’ needs. Learning is defined in
this research as the acquisition o f new knowledge that causes potential changes in
performance (Curtis et al., 1988; Huber, 1991).
There are a number o f overlapping constructs and terms to disentangle in
relationship to mental models and cognitive theories o f learning (Vandenbosch &
Higgins, 1996). The terms cognitive map (Tolman, 1959), schema (Bartlett, 1932),
and mental model (Craik, 1943) are widely used in psychology to explain the
workings o f human memory. “Mental models are considered indispensable to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
information processing because they organize knowledge in simple, robust and
parsimonious ways, in a world awash with information o f staggering complexity”
(Vandenbosch & Higgins, 1996, p. 200). Following Vandenbosch and Higgins
(1996), this study adopts the conceptualization o f learning as mental model
development.
Only when the knowledge received meets the knowledge seekers’ needs,
meaning that they can achieve desired mental model learning, can the knowledge be
perceived useful and valuable for future reuse (Markus, 2001). In the situations
where the knowledge received matches the learning objectives, knowledge seekers
are more motivated to process the advice and manage to learn (Wastell, 1999).
Consequently, the knowledge received is more likely to be perceived useful given
the match between the desired goals and the knowledge received. In a
cross-boundary context, due to the thought world barriers, knowledge holders may
have difficulties in understanding what knowledge seekers intend to learn. W ith a
request sent through e-mail, knowledge holders may interpret the problems
differently than the seeker’s thought world (Dougherty, 1992), and then provide
knowledge that will be perceived useless or irrelevant even if the argument is well
elaborated (i.e., high quality argument). While quality o f an argument m ay be an
important factor affecting knowledge usefulness, if the argument quality is not
relevant to a knowledge seeker’s intended learning, argument quality will have
minimal useful impact. This suggests that in a cross-boundary context, argument
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
quality cannot directly lead to perceived usefulness o f the knowledge received
through e-mail.
If the knowledge received cannot help knowledge seekers to achieve what
they intend to learn, anxiety may occur, which “alerts the organism to the presence
o f a significant mismatch between its desired goals and the current situation”
(Wastell, 1999, p. 583). As a result, the knowledge seeker m ay be reluctant to
process and accept the knowledge received, and engage in “defensive, antileaming
processes” (Wastell, 1999). Further, not only would the undesired help leave
knowledge seekers with their confusion and misconceptions intact, but it may have
had negative motivational effects as well, causing them to stop asking for help and
stop trying to understand (Webb & Palinscar, 1996). As such, knowledge seekers are
unlikely to perceive the knowledge received to be useful. Consequently, without
achieving desired learning, a well-organized advice (i.e., accurate, consistent, and
complete - high argument quality) cannot directly yield perceived usefulness o f the
knowledge received through e-mail. This leads to the following hypothesis:
Hypothesis 1: In the process o f e-mail-mediated knowledge transfer across
boundaries, achievement o f intended mental model learning mediates the
relationship between argument quality and perceived usefulness o f the knowledge
received through e-mail.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
3.3.2 Types o f Learning Intentions
Two types o f learning are considered in this study: M odel building and
Model maintenance. Vandenbosch and Higgins (1996) use N orm an’s theory to
■ y
identify two types o f mental model learning . M odel building (MB) involves the
formation o f “new conceptual structures” and is defined as the restructuring o f a
person’s cognitive model to adapt to new knowledge, a new cognitive framework or
anew perspective (Webb & Palincsar, 1996; Zhong & Majchrzak, 2004). The second
type is mental model maintenance. M odel maintenance (MM) describes the situation
in which “existing mental models are appropriate, or believed to be appropriate to a
given situation', learning occurs, then, when new information fits into the existing
mental models, and confirms what was already held to be true” (Vandenbosch and
Higgins, 1996, p. 202). To accomplish different tasks, knowledge seekers may intend
to achieve different learning and choose to conduct different search behavior (e.g.,
focused search vs. scanning). This dissertation is not specifically focused on
predicting this choice, but on examining the impacts o f this choice once made.
Cognitive research suggests that to achieve model building requires the
synthesis and restructuring o f the person’s perspectives on goals, interests, beliefs,
assumptions, business, and technical knowledge. In comparison, to achieve model
maintenance requires knowledge accumulation, verification, and confirmation to an
2 Norman (1982) identifies three modes of learning: accretion, structuring, and tuning. Vandenbosch and Higgins
(1996) compare the terminology used by prior researchers and adopt two modes o f mental model learning.
Tuning is not adopted because Norman’s tuning is not likely to be facilitated by executive support systems,
which was the focus in Vandenbosch and Higgins’s (1996) study. In this study, tuning is not likely to be
facilitated by e-mail-mediated advice seeking as tuning is “the fine adjustment o f knowledge to a task”, which
“may take thousands of hours of practice” to accomplish (Norman, 1982, pp. 81-82). Thus, tuning is not
investigated in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
existing mental model according to a given situation. Thus, MB is associated with
the reconceptualization and/or restructuring o f a knowledge seeker’s cognitive
mental model; whereas, M M is associated with the confirmation or verification o f a
knowledge seekers’ existing cognitive mental model. Obviously, “model building is
much riskier and its benefits more remote” (Vandenbosch & Higgins, 1996, p. 202).
Examining learning in the executive support systems context, Vandenbosch and
Higgins (1996) propose that the type o f learning achieved is determined by the way
how information is searched and acquired. That is, focused search (i.e., key
word-indexing in M arkus (2001)) makes mental model maintenance more likely the
result o f learning, and scanning (i.e., browsing information (Markus, 2001) without a
particular question to answer) is likely to lead to mental model building. This
suggests that the advice-seeking request should be defined differently to facilitate
different search behavior and learning needs. In the following sections, theory o f
knowledge reuse is discussed regarding how to define the advice-seeking request
properly to facilitate achievement o f knowledge seekers’ intended learning.
3.4 Defining the Advice-Seeking Request to Meet Learning Intentions
Due to the thought world barriers, knowledge holders m ay not understand
what knowledge seekers intend to leam (with a given question) and then provide
knowledge that w ill be perceived useless or irrelevant. A key facilitation activity is
to help knowledge seekers define their advice-seeking requests (Markus, 2001).
Knowledge seekers m ay not know the right question to ask when they send e-mail
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
solicitations across boundaries (Markus. 2001). The more dissimilar the knowledge
seekers are from knowledge holders (i.e., more boundaries to cross), the more
difficulty they m ay have in defining the advice-seeking request (Markus, 2001). As a
result, the more unlikely knowledge holders w ill understand the search intentions or
produce knowledge that will contribute to meet knowledge seekers’ learning needs.
Markus (2001) further provides insights on defining the advice-seeking
request. First, knowledge seekers should provide the background information (i.e.,
situational context) surrounding the question. As such, knowledge holders will
understand the target “symptoms” and present advice helpful but not conflicting or
irrelevant data. Second, knowledge seekers should describe their explication of
learning intent, and so knowledge holders will be able to provide necessary advice
(i.e., with or without alternate perspectives). Third, knowledge seekers should
include emotional incentive manifested as emotional inform ation to motivate
knowledge holders to respond. The following paragraphs describe how defining the
search question in this manner bridges the barriers to cross-departmental knowledge
access and positively influence achievement o f intended learning.
3.4.1 Richness o f Situational Context Information
Knowledge seekers do not always get what they need from the knowledge
received for reasons that have to do in part with how the knowledge is created
(Markus, 2001). For instance, some knowledge seekers m ay need to recontextualize
the knowledge received for their unique settings (Markus, 2001). In order for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
knowledge holders to provide relevant advice, knowledge seekers should provide the
situational context information in their request sent to e-mail distribution lists. This
way, knowledge holders are able to understand the right symptom w ith the situation
that knowledge seekers are operating and then, they are able to contribute to meet
knowledge seekers’ needs. In particular, situational context is essential for achieving
mental model development during electronic knowledge transfer. Underlying the
media richness theory (Daft & Lengel, 1986) is the assumption that e-mail makes it
harder to share localized knowledge between knowledge workers, because such a
medium “filters out significant cues, tends to be impersonal, and incurs a reduction
in language variety” (Lee, 1994, p. 145). Nevertheless, the barriers o f understanding
localized context involved in using e-mail would not necessarily be a problem if they
were balanced by appropriated social context specified in e-mail solicitations (Lee,
1994; Markus, 1994).
3.4.1.1 Importance o f Specifying Situational Context
It is crucial that any communicative act occurring in knowledge sharing rests
on a base o f mutual knowledge o f context. Nonaka et al. (2000) suggest that
organization could be conceptualized as a dynamic configuration o f “ba”, which is
defined as the context shared by those who interact with each other. And, “ba” is the
place where they create, share and use knowledge. Taylor (1996) calls it Information
Use Environments (IUE). IUE refers to “the set o f those elements that (a) affect the
flow and use of information messages into, within, and out o f any definable entity;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
and (b) determine the criteria by which the value o f information messages will be
judged” (Taylor, 1986, pp. 25-26). Furthermore, context is usually constructed
through layers around the core message that explains various assumptions (Te’eni,
2001).
Generally speaking, the situational context corresponds to “index
component” (Girin, 1990, p. 58, cited in Cossettee, 1998) in the analysis o f language.
According to Girin, this component is composed o f highly visible situational
elements that can be grouped under three broad headings: participants, places, and
time (Cossettee, 1998). Specifically, Situational context refers to information about
the communicators, the place o f the interaction, and time o f the interaction.
However, these elements must first be ascribed a meaning o f their own before they
can influence the language or the meaning ascribed to the statements of one of the
knowledge workers. As regards the “participants” group, the initial question is:
“Who is speaking to whom?” A person’s language and interpretation o f the
statements o f another person might be strongly influenced, for example, by the
meaning he or she ascribes to the fact that the other person is from a different
department and/or specializes in a different area (i.e., immediately-linked elements to
“who”). Each o f these elements (i.e., who, where, when) are immediately linked to
some other elements (Boland et al., 1994) as shown in the above example. If such
immediately-linked elements are also specified when a knowledge seeker specifies
situational context, it is considered interpretive context. Factual and interpretive
context refers to the contextual information including facts about who/where/when
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
coupled w ith the seeker’s own interpretation and knowledge o f these facts. In
contrast, if situational context is specified without corresponding explanations to
each o f these elements (i.e., the communicators, the place and time o f the interaction)
but facts only, it is referred to as factual context.
An example is discussed briefly to address the difference between these two
types o f situational context information. W hen describing situational context, a
knowledge seeker may phrase it this way, “Our Company has been asked to suggest
a panelist to present effective financial instruments”. This statement provides general
information representing factual context. In comparison, the seeker may specify
some immediately-linked elements this way: “The FYZ [the name o f an
organization] is organizing a rural finance workshop to be held M ay 23rd, and our
company has been asked to suggest a panelist to present effective financial
instruments for expanding rural financial services”. The latter description o f
situational context information provided interpretations to who (although it is “our
company”, it is actually from FYZ so knowledge holders would have a better idea of
the situation) and when (the workshop will be held on M ay 23rd, and so any advice
should be sent at least a few days before that particular date). Therefore, the
situational context specified in the latter example is considered factual and
interpretive context.
Reproduced with peim ission of the copyiight ownei. Furthei reproduction prohibited without peim ission.
53
3.4.1.2 Effect o f Situational Context
According to the definition o f learning, model maintenance should be
developed based on a given situation. The environment used as interpretive context
by a knowledge seeker is not a general one (i.e., summarized context), which can be
uniformly shared among knowledge workers, but is a personally constructed one
(Weick, 1991; Boland, 1993). Thus, to ensure the advice received be assimilated into
the existing mental models, factual and interpretive context is essential for
knowledge holders to offer targeted help by confining the advice in similar situations
- unique settings for knowledge seekers (Markus, 2001). W ithout a description of
immediately-linked elements to participants, place, and time (i.e., detailed situational
context information), knowledge holders may not be able to obtain a precise
understanding o f the circumstance embedded in knowledge seekers’ mental models -
the unique settings. Further, they might interpret the subject problem into a different
situation than the knowledge seekers’ (Dougherty, 1992). The advice, which cannot
fit into the given unique situation, will then be provided by knowledge holders based
on different situational assumptions (due to vague situational context information).
Such advice will not help knowledge seekers to incorporate the received knowledge
into their exiting mental models. Instead, they might view the knowledge received
irrelevant and stop processing it. Hence, it is unlikely the seekers can achieve their
intended model maintenance learning or judge the knowledge received as useful.
In contrast, model building may not require as detailed situational context as
model maintenance does. Model building learning requires changes and/or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
reconstruction o f the knowledge seeker’s existing mental model (Norman, 1982;
Vandenbosch & Higgins, 1996). Gaining substantial and extensive benefits from
different perspectives in other domains or in different situations is valuable for
achieving intended model building learning. Therefore, knowledge holders may not
need to understand the details of the unique settings o f the seekers, but provide
contextualized knowledge for the seekers’ learning purpose (Markus, 2001). It is
plausible that factual only context is sufficient to achieve model building.
Knowledge holders are not required to restrict their advice into a given or certain
circumstance (which is limited by immediately-linked elements to participants,
place, and time). Factual context information w ill allow knowledge holders to bring
more alternatives and/or sources in their cognitive systems. Factual (vs. interpretive)
situational context allows a self-diagnostic capacity for knowledge holders to be
aware of the perspectives from which knowledge seekers are operating, and “the
capacity to change one’s [knowledge seeker’s] point o f view”; and therefore “to
explore one’s [knowledge seeker’s] situation through a different light” (Smith, 1984,
p. 290). Interpretive situational context information w ill be extra, and limit the
abilities o f knowledge holders to offer multiple perspectives and/or different
alternatives, which are critical for achieving m odel building learning. Thus,
H2: In the process o f electronic knowledge transfer across boundaries, factual and
interpretive context specified in advice-seeking requests by knowledge seekers
positively affects their achievement o f intended model maintenance learning.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
H2a: In the process o f electronic knowledge transfer across boundaries, factual
context specified in advice-seeking requests by knowledge seekers positively affects
their achievement o f intended model building learning.
3.4.2 Explication o f Learning Intent
Markus (2001) distinguishes different knowledge reuse purposes, such as
acquiring new knowledge to handle a particular type o f problem, seeking answers to
create new knowledge and so forth. To meet these different objectives, some
knowledge seekers might conduct key word-indexing search, and other might
conduct document browsing (Markus, 2001). This is consistent with the learning
literature in terms o f the association between learning needs and search behavior
(Vandenbosch & Higgins, 1996). Knowledge seekers’ specification o f their search
intentions can be the resources for knowledge holders to identify knowledge seekers’
needs with reduced effort, and then the holders are more motivated to provide high
quality advice (Markus, 2001).
3.4.2.1 Importance o f Explication o f Learning Intent
Factors such as the knowledge seekers’ objectives or intentions may
considerably influence the interpretation o f the message (Cossette, 1998). The
“meaning” for the knowledge holder is the result o f a complex process o f trying to
understand what the knowledge seeker is saying through the e-mail message and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
why. Here again, the knowledge holder’s perception m ay be revised as a result o f the
meaning ascribed to the knowledge seeker’s statements o f two different intentions.
M arkus (2001) suggests different search behavior (i.e., key word-indexing
and document browsing) associated with different learning needs. In the same vein,
Vandenbosch and Higgins (1996) propose that the achieved learning is associated
with two ways o f conducting knowledge search: scanning and focused search.
Scanning has to do with the situation that people browse through information
without a particular problem to solve or question to answer; whereas, focused search
occurs when people are looking for specific information and/or verification
(Vandenbosch & Higgins, 1996). According to Vandenbosch and Higgins (1996),
focused search is clearly linked to mental model maintenance. Also, the relationship
between scanning and metal model building is substantially stronger than the
relationship between scanning and mental model maintenance. Given such different
knowledge transfer needs, knowledge seekers should convey their learning
objectives or search strategies to knowledge holders to shift the burden o f packaging
knowledge from knowledge holders (Markus, 2001). This w ill also provide
incentives for the knowledge holders to make contributions because they have
reduced cost associated with contributing (Markus, 2001).
Explication o f learning intent can be described in terms o f the following two
dimensions. Explication o f MB intent refers to the situation that knowledge seekers
intend to conduct scanning, or hope that knowledge holders would provide various
perspectives (and/or multiple interpretations, alternatives) enabling knowledge
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
seekers to recognize different ways or new ideas to refine their cognitive structure
(i.e., open to all available alternatives when conducting scanning). In contrast,
explication o f M M intent refers to the situation that knowledge seekers intend to
conduct focused search, or hope that knowledge holders would provide confirmation
or verification o f knowledge seekers’ existing perspectives without providing
different interpretations (i.e., not open to various alternatives).
3.4.2.2 Effect o f Explication o f learning intent
Explication o f learning intent deals with knowledge seekers’ learning
intentions and/or the purpose o f sending out e-mail request for help. W ith specified
explication o f learning intent in e-mail messages, knowledge holders can be more
purposeful to contribute to either model building or model maintenance learning as
they know the knowledge seekers’ learning needs. In other words, resources o f how
to package the knowledge are provided to knowledge holders for them to provide
high quality targeted advice (Markus, 2001).
If knowledge seekers intend to achieve model building learning, they need to
express their MB intent in the e-mail solicitation. In information systems
development (ISD), this form o f learning is described as clients and developers
creating “a new domain view” (Kirsch & Beath, 1996, p. 224). This new domain
view involves the synthesis and restructuring o f every knowledge w orker’s
perspective on goals, interests, beliefs, assumptions, business, and technical
knowledge. Only when knowledge seekers elaborate their intentions o f being open to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
diverse alternatives, can knowledge holders package their knowledge to provide
different perspectives, new knowledge and information, which m ay induce seekers’
knowledge integration (i.e., model building) (Grant, 1996). Cognitive theorists (see
reviews by W ebb & Palincsar, 1996; and O ’Donnell & O ’Kelly, 1994) have studied
cooperative learning groups and concluded that groups are more likely to offer
opportunities to change participants’ cognitive structures when they apply the
approach called “cognitive elaboration”. Explication o f MB intent m ay stimulate
potential knowledge holders to practice cognitive elaboration, and thereafter,
knowledge seekers are more likely to achieve intended model building learning. This
suggests that high explication o f learning intent specified by knowledge seekers
should facilitate knowledge holders to bring new perspectives and
conceptualizations, and further facilitate achieving intended model building learning.
For instance, “If anyone has a simple business diagnostic tool for guidelines for
business analysis, we would appreciate your perspectives” implies that the seeker is
open to all perspectives, new ideas, or alternatives. This way, the potential
knowledge holders may be able to provide advice to contribute to rearranging or
redefining the seeker’s mental models by interpreting and incorporating new
information (Vandenbosch & Higgins, 1996).
In comparison, model maintenance learning requires specification o f
explication o f MM intent in knowledge seekers’ e-mail solicitations. Model
maintenance learning focuses on knowledge accumulation to fit into existing
cognitive structures (Vandenbosch & Higgins, 1996). W hen knowledge seekers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
intend to gain verification or confirmation o f their existing perspectives, they must
explicate this intention (i.e., focused search) in their e-mail requests. With such
information, knowledge holders would understand they should not provide any other
perspectives and/or alternatives (which might be perceived useless). The expression
o f low explication o f learning intent serves as the resources for knowledge holders to
produce high quality advice for consumption by knowledge seekers (Markus, 2001).
One reason is that the cost and effort associated is predictable; and another reason is
that knowledge holders have the direction to provide advice that m ay meet
knowledge seekers’ learning needs. Therefore, knowledge seekers should include in
their e-mail solicitations that they are not open to substantial alternatives and/or new
perspectives. Thus,
H3: In the process o f electronic knowledge transfer across boundaries, explication
o f M M intent specified in advice-seeking requests by knowledge seekers positively
affects their achievement o f intended model maintenance learning.
H3a: In the process o f electronic knowledge transfer across boundaries, explication
o f MB intent specified in advice-seeking requests by knowledge seekers positively
affects their achievement o f intended model building learning.
3.4.3 Emotional Incentive
An inhibitor to quality contributions (that support knowledge seekers’
achieving intended learning) is the lack of incentive to provide high quality, sanitized
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
knowledge (Markus, 2001). The influence o f a knowledge holder’s motivation in the
process o f knowledge transfer has been documented by several scholars (e.g.,
Szulanski, 1996). Increasing knowledge holders’ incentives to contribute is critical in
successful knowledge transfer processes (Markus, 2001).
3.4.3.1 Importance o f Providing Emotional Incentive
Knowledge holder incentive is essential in knowledge transfer processes
(Markus, 2001). The costs involved in creating knowledge and producing advice
would not necessarily be a problem if they were balanced by appropriate incentives
(Markus, 2001). W ithout incentives or motivations to contribute, knowledge holders
are often “unwilling to devote time and resources to support the transfer” (Szulanski
1996, p. 31). W ithout an extrinsic reward system in e-mail-mediated knowledge
transfer processes, it is very important for knowledge seekers to motivate knowledge
holders to contribute in the e-mail solicitations.
Emotional incentive refers to the description o f attitude or feelings which one
communicator has about the issue communicated to attract the receivers’ attention to
respond, and can be part of the emotional information communicated (Schwarz,
1990). The emotional factors will influence to some extent both the language used
and the meaning ascribed by one communicator to the statements made by the other
(Te’eni, 2001). If the emotional information is merely described (in terms o f fear,
hesitation, admiration, urgency, and so forth) without the provision o f reasons or
explanations, it is considered low level emotional incentive. If the underlying reasons
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
for the feelings are explained to call for knowledge holders’ attention and motivate
them to contribute, it is viewed as high level emotional incentive.
3.4.3.2 Effect o f Emotional Incentive
Emotional incentive is manifested as the emotional information provided in
the e-mail solicitations. Extensive research clearly shows that emotion has a major
impact on social cognition, judgments, and reasoning (Damasio, 1994; Oatley &
Jenkins, 1996). Emotional influences occur because social thinking necessarily
involves highly constructive, generative mental operations (Forgas, 2000). A
multi-process theory, the Affect Infusion Model (AIM), explains the effects of
emotional context on information processing and learning. Affect infusion refers to
“the process whereby affectively loaded information exerts an influence on, and
becomes incorporated into cognitive and judgm ental processes, entering into a
person’s deliberations and eventually coloring the outcomes” (Forgas, 2000, p. 255).
The basic assumption behind AIM is that affect (affect is interchangeable with
emotion in this study) can have both informational and processing effects on
cognition.
M odel maintenance learning requires knowledge holders to verify or confirm
knowledge seekers’ existing perspectives, which involves relatively less cognitive
effort than model building learning. As such, the cost associated with providing high
quality advice might be less than model building learning. Hence, provision o f brief
emotional information (without detailed underlying reasons or explanations) to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
motivate knowledge holders to respond m ight be sufficient incentives. In contrast,
model building learning incurs more constructive and open-ended processing and
information search strategies; and thus facilitates affect infusion (Clore et al., 1994).
The AIM predicts that affect infusion is m ost likely to occur whenever circumstances
promote elaborative, open, constructive information processing style (Fiedler, 1991;
Forgas, 1991). This indicates that the situation when knowledge holders need to
contribute to model building learning promotes affect infusion. Further, the cost
increases when knowledge holders anticipate increased cognitive effort to produce
high quality advice so as to facilitate model building learning (Markus, 2001). In
order to balance the increased cost, underlying reasons for the emotional information
should be specified to attract knowledge holders’ attention and motivate them to
contribute. As a result, it is plausible that achieving model building learning requires
detailed emotional incentive than model maintenance. This leads to the following
hypotheses,
H4: In the process o f electronic knowledge transfer across boundaries, brief
emotional incentive specified in advice-seeking requests by knowledge seekers
positively affects their achievement o f intended model maintenance learning.
H4a: In the process o f electronic knowledge transfer across boundaries, detailed
emotional incentive specified in advice-seeking requests by knowledge seekers
positively affects their achievement o f intended model building learning.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
3.4.4 Effect o f Boundary
“Learning organizations” are celebrated for the ability to share knowledge
from both internal and external sources o f knowledge (Nonaka & Takeuchi, 1995).
Along the same line, Rosenkopf and Nerkar (2001) claimed the importance of
considering organizational boundaries between firms when we study knowledge
transfer. The empirical evidence from Stuart and Podolny (1996) suggests that
knowledge workers across organizations are likely to focus their knowledge
exploration on closely related technological domains. Drawing from the theory o f
thought worlds, different funds o f knowledge, systems o f meaning, and readiness for
directed perception create barriers to accessing cross-departmental knowledge
(Dougerhety, 1992). Boundaries essentially reflect thought world barriers (i.e., the
extent o f differences between thought worlds). The more boundaries, the more
barriers, and then the more difficulty in defining the search question (Markus, 2001).
This suggests that boundary may influence the effect o f the search question on the
achievement o f intended learning.
Rosenkopf and Nerkar (2001) highlight the difference o f knowledge
exploration by distinguishing different boundaries (i.e., technological boundary
spanning and organizational boundary spanning). Likewise, knowledge exploration
within organizations requires technical area boundary spanning (i.e., specialty) and
unit (i.e., department) boundary spanning. Knowledge seeking through e-mail
systems transcends space and time, and further, requires knowledge exploration to
transcend diverse boundaries (i.e., both unit boundary and specialty boundary, unit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
boundary only, specialty boundary only, or no boundary) (Dougherty, 1992). The
more boundaries that need to be crossed, the more radical changes in meaning may
occur (Rosenkopf & Nerkar, 2001); and therefore, the more important it is to define
a good search question to mitigate communication complexities, explicate intentions,
and reduce ambiguities (Markus, 2001). Thus,
H5: During the process o f electronic knowledge transfer, organizational
boundaries moderate the relationship between the features o f defining
advice-seeking requests (i.e., specification o f situational context, explication o f
learning intent, emotional incentive) and achieving intended mental model learning.
3.5 Control Variables
Research indicates that knowledge holder credibility m ay be relevant both
within the context o f CMC and within knowledge work more generally construed. In
the context o f text-based messages, holder credibility describes the degree to which a
knowledge seeker believes a potential knowledge holder is credible (e.g., competent,
trustworthy, reliable), reflecting nothing o f the message itself (Sussman & Siegal,
2003). Individually, people often use cues pertaining to the m essage’s source when
they are unable or unwilling to expend the effort to elaborate on the message’s
context (Sussman & Siegal, 2003). Also, research suggests that individuals’
perceptions o f knowledge usefulness is influenced by the extent they believe that
knowledge holder is credible (Markus, 2001). Early laboratory experiments on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
role o f credibility found significantly more opinion change in the direction advocated
by the communicator when the material was attributed to a high-credibility source
than it was attributed to a low-credibility source (Hovland et al., 1953). As such,
knowledge holder credibility is expected to positively relate to perceived usefulness
o f the knowledge received through e-mail solicitation.
Prior expertise o f knowledge seekers alters their perceptions of knowledge
usefulness by affecting individuals’ ability to process (Markus, 2001). Higher levels
o f seekers’ prior knowledge and comprehension about the message topic increase the
quantity and depth o f issue-relevant thoughts than occur to them and their ability to
understand them (Sussman & Siegal, 2003). Research suggests that individuals who
are non-experts are confronted with an intellectual challenge (Amabile et al., 1996;
Oldham & Cummings, 1996) and m ay lack suitable criteria forjudging the
knowledge received (Markus, 2001). A knowledge seeker’s expertise influences how
they judge the knowledge that they receive. Thus, knowledge seeker’s expertise is a
possible contributor to perceived usefulness o f the knowledge received.
Knowledge seekers’ involvement levels also alter their perceptions of
knowledge usefulness. High levels o f seekers’ involvement tend to motivate
increased interest in the received knowledge (Sussman & Siegal, 2003). Low levels
o f seekers’ involvement may prevent them from processing the received knowledge
(Szulanski, 1996), and the knowledge is less likely to be perceived useful.
Consequently, in this dissertation study, knowledge seekers’ involvement is
controlled as it might contribute to perceived usefulness o f the knowledge received.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
3.6 Conceptual Model
Using the theory o f knowledge reuse and the theory o f thought worlds as
discussed earlier in this chapter, the conceptual model is shown in Figure 3.1.
Figure 3.1 The Hypotheses and Proposed M odel
H2a H3a
Boundary -Hfr
Argum ent H 1
Quality
Perceived U sefulness
o f K now ledge
R eceived
H4
Boundary
H3
H2
Control Variable
- Holder credibility
- Seekers’ expertise
-.-Seekers’ involvem ei
Factual Context Explication of
MB intent
Explication of MM
intent
Brief Emotional
Incentive
Detailed Emotional
Incentive
Factual and
Interpretive Context
Achievement of Intended Model
Building Learning
Achievement of Intended Model
Maintenance Learning_________
The model can be broadly described as follows. A chievem ent of intended
mental model learning mediates the relationship between argument quality and
perceived usefulness o f the knowledge received. Proper provision o f three elements
in the advice-seeking requests helps knowledge seekers achieve intended mental
model learning. The better the three features deployed in e-mail requests by
knowledge seekers, the more likely their intended mental m odel learning will be
achieved. Consequently, the knowledge received is more likely to be perceived as
useful. In particular, factual context, explication o f MM intent, and brief emotional
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
incentive are associated with achieving intended model maintenance learning.
Interpretive and factual context, explication o f MB intent, and detailed emotional
incentive are associated with achieving intended model building learning. Finally, as
more exploration boundaries are being crossed, it is more essential to employ the
three dimensions in defining the advice-seeking request in electronic solicitations.
Three control variables are considered in the model: Knowledge holder credibility,
Knowledge seeker’s involvement, and seeker’s expertise.
3.7 Summary
This dissertation study draws from the theory o f thought worlds and M arkus’
theory o f knowledge reuse to propose a set o f strategies (i.e., three features to be
deployed in defining the advice-seeking request) that support achievement o f
intended mental model learning. It is first proposed that the relationship between
argument quality and perceived usefulness o f the knowledge received is mediated by
achievement o f intended mental model learning in e-mail-facilitated cross-boundary
knowledge transfer. The relationship is proposed with controlling for other factors
found to be important in previous research on e-mail-facilitated knowledge transfer
including: knowledge holders’ credibility, and knowledge seekers’ expertise and
involvement. Further, it is hypothesized that three dimensions o f advice-seeking
definition (i.e., situational context, explication o f learning intent, and emotional
incentive) positively affect achievement o f intended mental model learning.
Moreover, knowledge exploration boundary is the moderator in the relationship
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
between each specific dim ension o f the advice-seeking request and achievement of
intended learning.
The next chapter explains the research methods. It begins w ith an introduction
to the research setting at Company A. The data collection activities (including
qualitative and quantitative approaches) are discussed as well. In addition, the
two-phase survey administration to Company A employees is described, including
the operationalization o f the variables under study. Coding methodologies and
examples are then discussed. The rationale behind using Partial Least Squares (PLS)
as the primary data analytic technique is also provided.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
CHAPTER 4: RESEARCH METHODS
To test the hypotheses, a two-phase approach was deployed including a
qualitative study (including onsite observations, interviews, and reviews o f e-mail
examples) and a quantitative study (i.e., field survey). Conducting the research at an
organization where cross-boundaries knowledge seeking was important to
successfully completing employees’ work helped capture the complexity o f the
phenomenon. Company A was chosen as the data collection site (section 4.1). To
ensure that the w ay employees at Company A sought advice through e-mail was
accurately incorporated into the survey, a qualitative study was conducted first
(section 4.2). For 8 months prior to the quantitative study, observations o f e-mail
solicitation, interviews with employees, and reviews o f the e-mail samples provided
by the informants from Company A were used to confirm the appropriateness o f the
constructs selected for the conceptual model and contextualize the survey.
Quantitative study including survey and coding methodologies is then discussed
(section 4.3). All surveys were obtained from a single site to control for the
organizational effect and the basic nature o f how the work was performed. Section
4.3.1 describes the survey methodology, including the survey instruments employed.
Section 4.3.2 discusses coding methodology. The use o f Partial Least Squares (PLS)
as the primary data analytic technique is also discussed (section 4.4).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
4.1 Research Setting
The research informants consist o f employees in a multi-national firm
(hereafter called Company A). Company A is a global consulting firm promoting
economic growth and higher living standards in developing countries. The firm has
its headquarters in the US and more than 60 field offices worldwide focusing on
different projects (e.g., A fghan’s econom y rebuilt; El Salvador coffee business;
Opening microfmance opportunities in Sudan; etc). W ith offices on five continents
and close to 2,000 employees, it is one o f the world's largest firms focusing
exclusively on international development. Projects are undertaken based on
geographical regions that different departments focus on (i.e., Africa, Asia, Europe
and Eurasia, Latin America and Caribbean, and M iddle East). In order to better
facilitate various projects all over the world, Company A has recently built up their
Practice Network system (e.g., Agriculture/Agribusiness; Democracy & Governance;
Finance and Banking; etc) across regions. Employees at Company A are encouraged
to join the Practice Network e-mail groups that they are interested in or specialized
in.
The employees at Company A are extremely reliant on e-mail for getting
their work done. One major purpose o f these e-mails is to seek work-related advice,
which contains recommendations and/or suggestions on how to solve specific
problems, when transferred to knowledge seekers. Respondents were selected from
professionals who have recently sent out e-mail solicitation and acquired advice from
other knowledge workers through e-mails. Several e-mail examples are presented in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
Table 4.1 to illustrate the type o f advice-seeking message collected in this study. All
names and identifying information have been altered to protect confidentiality.
Table 4.1 Examples of Advice-seeking E-mail
________________________Advice-seeking E-mail Messages________________________
Roma: METD has just opened the METD North Office in XYZ and is currently
working intensively with the staff o f the GIO to strengthen their ability to conduct
business diagnostics - from interviewing techniques, to identifying immediate
business needs and providing practical, demand-driven solutions to these needs. If
anyone has a simple business diagnostic tool or guidelines for business analysis o f
this type we would appreciate hearing from you..._______________________________
Irene: Do we have any precedent whereby a guarantee fund has been retained by a
subcontracted bank beyond a TR project end date? If so, do w e have examples
whereby any XYZ M ission has accepted this or a similar method?_________________
4.2 Qualitative Study
Qualitative study methods included observations o f e-mail solicitations and
responses, interviews with employees, and reviews o f the e-mail samples provided
by the informants.
During Spring 2 0 0 4 ,1 spent two months at Company A learning about its
knowledge-sharing practices through its e-mail systems. Seventy-two people were
interviewed during that time, with interviews ranging from 5 minutes to 90 minutes.
The interviewees included Project Administrators (PA), A ssistant Project
Administrators (APA), Project Managers (PM), and Senior Vice Presidents (SVP).
Questions were asked about the experience o f seeking advice and providing advice
via e-mail. From these interviews, I learned that the nature o f Company A
employees’ work makes them extremely dependent on e-mail to perform their
day-to-day work, o f which one major purpose o f these e-mail exchanges in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
company is to seek work-related advice. The process o f e-mail-mediated advice
seeking across boundaries required interpretations from different specialties. Given
the excellent knowledge sharing culture throughout the company, knowledge seekers
usually receive m any responses to a posted request. However, what I discovered was
that m any e-mail responses were considered irrelevant and/or useless. The scenario
(i.e., Tom & Jane’s story) provided at the onset o f this dissertation is an example.
During these qualitative interviews, I probed for reasons for the irrelevant
sharing o f advice across divisions and specialties. One o f the reasons mentioned was
the lack o f sharing adequate information about knowledge seekers’ objectives and
learning needs. E-mail requests that included clearly specified intentions and
contexts seemed to attract more useful and relevant advice from knowledge holders.
This led to several e-mail interactions until knowledge holders finally understood the
underlying assumptions of the questions and the intentions o f the knowledge seekers.
Then, knowledge holders were able to help, as in Tom & Jane’s example.
The interviews indicated that many people stopped responding to requests
for advice, fearing the huge amount o f time and effort involved in figuring out what
information the knowledge seekers wanted. In addition, without sufficient
information to clearly identify the expectations surrounding an e-mail request, senior
knowledge holders had so many resources at their fingertips that they often were
unsure as to which o f the many resources available to them they should share with
the knowledge seekers. Further, without an appropriately-defined advice-seeking
request in the e-mail to indicate the situation and intention considered by the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
knowledge seeker, potential knowledge holders did not know w hether their advice
was going to be helpful (i.e., lack o f resource to contribute, M arkus (2001)). As a
result, knowledge seekers reported often receiving e-mails from knowledge holders
that contained irrelevant information or suggestions (i.e., not useful for what
knowledge seekers intend to learn). For all o f these reasons, informants reported
that knowledge sharing via e-mail among Company A ’s employees was ineffective.
In fact, 89% o f the informants complained that e-mail traffic was not leading to
useful knowledge.
The reviews o f e-mail examples also yielded some good examples of
effective advice-seeking. The single e-mail request demonstrated in Table 4.2
outlined a well-defined advice-seeking request, which yielded responses the seeker
thought very helpful. More specifically, the knowledge seeker provided factual
situational context information (i.e., general background inform ation o f the
solicitation with visible situational elements), high explication o f learning intent (i.e.,
the seeker’s hopes as regards the knowledge holder’s ability to understand the
message made and the expectations o f diverse alternatives), and high level emotional
incentive (i.e., the feelings o f the knowledge seeker with reasons and explanations to
that feeling). W ith such a well-defined advice-seeking request, the knowledge holder
stated that he was able to understand the situation and purpose correctly, interpret the
solicitation message properly, and react to the seeker with helpful suggestions. As a
consequence, the knowledge seeker felt he received very useful and relevant advice,
and achieved what he intended to learn (model building learning in this case).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
Table 4.2 An Example o f a W ell-defined Advice-seeking Request with Proper
Dimensions
Content in the e-mail request Dimension o f the
advice-seeking request
I'm sorry to inundate your inboxes once again, but this
issue has come up again with regards to the fielding o f
John and Samantha. Sunny and I are both having trouble
convincing Accounting to issue travel/ABC advances
Factual situational
context information
Sunny is going to talk to Louis about the issue at hand,
but that doesn't necessarily fix the problem as it will
surface again with Travel and anyone else we field down
the road. I’d like to get any o f your opinions or
experience associated with this issue, in different regions,
maybe. Does anyone have any suggestions as to how we
can proceed?
Explication o f MB
intent
I am very hesitant to further push the support units any
further because I feel that it undermines the integrity o f
the project management for which company A is
recognized. So it is very important now for you to help
me solve this recurring problem.
Detailed emotional
incentive
The qualitative study served as an initial step to understand the proposed
conceptual model in an organizational setting. A quantitative study is then discussed
to further test the hypotheses.
4.3 Quantitative Study
The quantitative study consisted o f a two-phase survey o f knowledge
seekers (section 4.3.1) and coding o f e-mail messages sent as requests (section
4.3.2). The survey was used to assess constructs o f argument quality, achievement of
intended learning (i.e., MB or MM learning), knowledge usefulness, and control
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
variables (section 4.3.1). Coding e-mails assessed the dimensions specified in e-mail
requests and boundary information (section 4.3.2).
4.3.1 Survey M ethodology
Detailed discussions with the Knowledge M anagem ent Research Steering
Committee in Company A (regarding the proposed research model and the
questionnaire) confirmed the suitability o f administering the questionnaire in
company A. Prior to the formal administration o f the survey, the survey was piloted
twice with students (i.e., 2 undergraduate students majored in M anagement
Information Systems and 6 graduate students specialized in Computer Science) and 8
times with Company A employees. Instructed to read the survey, answer each
question, and provide e-mail samples, students and employees were also encouraged
to question and comment on the clarity, appropriateness, and understandability o f
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 3 times until the survey instructions, questions, and items were considered
to be clear, appropriate, and well understood.
Next, the formal two-phase survey was administered to test the major
hypotheses under study. The first survey was administered when I observed one
e-mail solicitation; and the second survey was administered one week after (section
4.3.1.1). Two-phase survey was administered to assess the intended learning and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
actually achieved learning separately. All the survey instruments were adapted from
existing literature (section 4.3.1.2).
4.3.1.1 Survey Administration
Both the first survey and the second survey were sent out directly to the
potential respondents through e-mails. The two questionnaires were sent separately
in order to obtain intended learning when the e-mail was first sent out, and achieved
learning after responses were received.
The following criterion was established for identifying target respondents.
The potential respondents had to have sent out an e-mail solicitation to get advice for
a nonadministrative problem (excluding administrative questions, or simple file
transfer request). Advice refers to work-related advice, and should contain
recommendations and/or suggestions on how to solve specific problems, when
transferred to knowledge seekers (Sussman & Siegal, 2003). Several e-mail
examples are presented below (Table 4.3) to illustrate advice messages in this study.
All names and identifying information have been altered to protect confidentiality.
Table 4.3 Examples of Advice Received by E-mail
Advice received by e-mail
Samantha You m ay want to consider using X Y Z’s Development Credit
Authority (DCA) guarantee mechanism which will cover up to 50% o f
the risk o f SBCA’s partner bank for making longer term loans o f up to
5 years...
John You may want to try looking at the State Department's Trafficking in
Person's report, and the following organizations for sex trafficking
data: Anti-Slavery International; The Protection Project (out of John's
ABC University). W ith a bit more time I can come up with m ore...
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
The actual data collection was done from mid-November, 2004 to
mid-March, 2005. An introductory e-mail about this study was sent to all employees
at the Company signed by the Knowledge M anagem ent Division which consisted o f
Vice President. Also, I was added to all e-mail groups (e.g., regional e-mail groups,
practice network e-mail groups) in the Company to m onitor e-mail threads and given
permission to send two questionnaires to any company employees. From then on, as
long as I saw an e-mail solicitation sent out regarding a specific work-related
question that needed to be solved with advice, an e-mail with Questionnaire I was
sent to the knowledge seeker. W ith Questionnaire I, intended learning (i.e., MM or
MB learning), and control variables (i.e., knowledge seeker’s expertise, seeker’s
involvement, and knowledge holder’s credibility) were assessed. If no response was
received within the timeframe provided, phone calls were placed to the potential
respondents to confirm their receipt of the original e-mail with Questionnaire I, and
remind them o f the participation in this research study. A second reminder was
administered to those who had not submitted a survey after the first reminder. Those
individuals who did not respond after two reminders were no longer respondents. A
week later, an e-mail with Questionnaire II was sent to the respondents (who have
completed Questionnaire I) to assess the argument quality o f the most useful advice
they have received (or, the most recent advice if none o f the advice is useful),
knowledge usefulness o f the advice, achieved learning (i.e., MM or MB learning) out
o f the advice, and demographic information. Phone calls were placed if respondents
did not respond to Questionnaire II.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
During the four-month data collection process, 136 employees at the
company sent out e-mails asking for advice and were invited to complete the survey.
O f these 136 employees, 122 responded to Questionnaire I (with response rate
89.7%), and 117 responded to both Questionnaire I and Questionnaire II, yielding a
response rate o f 86%. Out o f these 117 responses, 8 were discarded for incomplete
responses, yielding 109 that were available for data analysis.
4.3.1.2 Survey Instruments
All constructs measured in the questionnaires have been measured in
previous research and thus, I used existent scales.
Argument Quality. A three-item scale developed by Bailey and Pearson (1983) and
validated by Sussman and Siegal (2003) was used to assess the quality o f the
argument in the advice received through e-mail. Knowledge seekers were asked to
enter a value from 1 to 7 representing: The information in this e-mail is complete (to
incomplete), consistent (to inconsistent), and accurate (to inaccurate). One factor was
expected and shown in the factor analysis (Table 4.4). The Cronbach’s alpha was
adequate at 0.813.
Mental Model Learning. A four-item scale was adapted from Vendenbosch and
Higgins (1996) to assess model building learning and model maintenance learning.
Cognitive science researchers often measure learning by testing for specific changes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
in content knowledge obtained (e.g., W illoughby et al., 2000). Nevertheless, in this
study, there was no preconceived idea o f what each knowledge seeker might have
intended to learn and actually learnt. Thus, knowledge seekers were expected to be
sufficiently self- reflective as to be able to report when they had “confirmed what
they thought” or “changed their minds”. Therefore, knowledge seekers were asked:
“Please tell us, overall, the extent to which the responding e-mail message led you
to: (1) Develop new ideas, (2) Gain new perspectives, (3) Cause you to change your
mind, (4) Gain new ways to do things, (5) M aintain your perspective on an issue, (6)
Reinforce your current beliefs, (7) Verify your assumptions, [and] (8) Justify your
decisions”. A two-factor solution was expected and shown in the factor analysis
(Table 4.4). The Cronbach’s alphas for model building learning and model
maintenance learning were acceptable at 0.862 and 0.921 respectively.
Questionnaire I was sent to assess desired mental model learning. One week later,
Questionnaire II was sent to assess actual mental model learning after knowledge
seekers received advice. Achievement o f intended learning is assessed by the
difference o f the scores between Questionnaire I and II.
Knowledge Usefulness. A three-item scale developed by Bailey and Pearson (1983)
and validated by Sussman and Siegal (2003) was used to assess the usefulness o f the
advice received through e-mail. Knowledge seekers w ere asked to what extend the
information in the received e-mail is valuable, informative, and helpful. One factor
was expected and shown in the factor analysis (Table 4.4). The Cronbach’s alpha
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
was adequate at 0.871. In sum, all items loaded cleanly on their original scale items.
Cronbach’s alpha for all multi-item measures was high and above 0.8 (Table 4.4).
Table 4.4 List o f items measuring the constructs and their Cronbach’s alpha
Construct Questions and Items Item loadings in
PLS output
Cronbach’s
alpha
Argument Quality
The information in this E-mail is
complete [ARGUQUA11
.78
.81
The information in this E-mail is
consistent [ARGUQUA21
.81
The information in this E-mail is
accurate [ARGUQUA31
.84
Mental
Model
Learning
(Qi)
Model
Building
Develop new ideas [MB 1 1 .77
.86
Gain new perspectives [MB21 .73
Cause you to change your mind
[MB31
.82
Gain new ways to do things [MB4] .83
Model
Maintenance
Maintain your perspective on an issue
[MM 1 1
.79
.91 Reinforce your current beliefs [MM2] .89
Verify your assumptions [MM31 .88
Justify your decisions [MM41 .95
Mental
Model
Learning
(Qii)
Model
Building
Develop new ideas [MB 1 1 .75
.82
Gain new perspectives [MB21 .87
Cause you to change your mind
[MB3]
.77
Gain new ways to do things [MB41 .78
Model
Maintenance
Maintain your perspective on an issue
[MM 1 1
.88
.85 Reinforce your current beliefs [MM21 .76
Verify your assumptions [MM31 .84
Justify your decisions [MM41 .79
Knowledge Usefulness
The information in this E-mail is
valuable [KNOWUSE11
.81
.87 The information in this E-mail is
informative [KNOWUSE21
.83
The information in this E-mail is
helpful [KNOWUSE31
.85
*p<0.05
**N= 109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
Control Variables The measures for control variables were adapted from the
existing literature. Sussman and Siegal’s (2003) index o f source credibility was
used to measure knowledge holder’s credibility. The knowledge seekers were asked:
“(1) How knowledgeable is the person who wrote this message, on the topic o f the
message (Please enter a value from 1 to 7 representing from 1 = N ot knowledgeable;
to 7 = Knowledgeable), [and] (2) To what extend is the person who wrote this
message an expert on the message topic (Please enter a value from 1 to 7
representing from 1 = N ot expert; to 7 = Expert)”. Sussman and Siegal’s (2003)
indices o f expertise and involvement were adapted and used to measure knowledge
seeker’s expertise and involvement respectively. Knowledge seekers’ expertise was
assessed by asking: “(1) How informed are you on the subject m atter o f this issue,
(2) To what extent are you an expert on the subject m atter o f this e-mail request”,
and (3) How familiar are you with the subject m atter o f this issue. An alpha o f 0.81
was obtained. Knowledge seekers’ involvement was measured by asking: “(1) How
involved are you in the subject matter o f this email, (2) How m uch has the issue
discussed in this email been on your mind lately”, and (3) How m uch has the issue
been discussed by you and your colleagues lately. An alpha o f 0.71 was obtained.
4.3.2 Coding M ethodology
This study focuses on written communication (especially e-mail), which
means not spoken, but rather typed and/or electronically transmitted text. A total o f
109 e-mail messages sent out by the knowledge seekers were coded. Coding
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
instructions were adapted from T e’eni et al. (2001) (for situational context
information), Vandenbosch and Higgins (1996) (for explication o f learning intent),
and Forgas (2000) (for emotional incentive). The e-mail received containing advice
was not coded. This study focuses on advice-seeking behavior in the very early stage
o f knowledge transfer - the effect o f advice-seeking request in e-mail solicitations on
achieving intended learning by knowledge seekers. Given this research purpose, only
the e-mail request was coded in order to investigate the impact o f dimensions of
advice-seeking requests provided by knowledge seekers. Coding e-mail received
containing advice would be interesting for additional studies to examine knowledge
holders’ perceptions o f the advice-seeking requests provided by seekers.
After a detailed instruction meeting, a trained coder analyzed several e-mail
messages according to the procedure detailed below (Table 4.5). The coder and the
author then met to compare the coding and resolve disagreements. Inter-rater
reliability was conducted to confirm the validity o f the coding instructions. A sample
o f 30 messages was coded by both the coder and the author to assess the inter-rater
reliability. The proportion o f cases for which the author and the coder agreed was
80%, which is acceptable. Table 4.6 presents some coding examples. Each o f the 109
e-mail messages was coded and later fed into PLS for analysis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.5 Coding Procedure and Instructions
83
Instructions
Step 1 Read the entire e-mail message and then record structured
parameters: sender, receiver(s), and receiver(s’) group(s). The
group information o f the receivers (i.e., which department and/or
which practice network that the receivers belong) will be
considered together with the group information o f the senders
(from survey) to identify the boundary level in the process o f
electronic knowledge seeking.
Step 2 W ithin an e-mail message, read one sentence at a time and treat it
as a basic unit to code according to the three dimensions o f
advice-seeking definition.
Step 3 Code message for articulation o f situational context: Situational
context information was defined as highly visible situational
elements that can be grouped under three broad headings:
participants, places, and time. Put 0 if such information is not
provided. Put 1 if such information is provided w ith facts only
(without explanations to instantly linked elements) (Factual
situational context). Put 2 if such information is provided including
facts about who/where/when coupled with the seeker’s own
interpretation and knowledge o f these facts (Factual and
Interpretive context).
Step 4 Code message for use o f explication o f learning intent: Explication
o f learning intent was defined as the descriptions o f knowledge
seekers’ hopes as regards the knowledge holders’ ability to
understand the messages made (e.g., the purpose o f the solicitation)
and/or seekers’ expectations o f the alternatives that w ill be
receiving. Put 0 if such information is not provided. Put 1 if such
information is provided with seekers’ particular intentions - to
receive verification and/or confirmation o f a specific circumstance
without alternatives (Explication o f MM intent). Put 2 if such
information is provided with seekers’ intentions - opening to
various advices, information, and/or alternatives without limitation
(Explication o f MB intent).
Step 5 Code message for use o f emotional incentive: Emotional incentive
refers to the description o f attitude or feelings which one
communicator has about the issue communicated to attract the
receivers’ attention to respond. Put 1 if emotional incentive is
described with reasons or explanations to the feelings outlined in
the request (Detailed emotional incentive). Put 0 if emotional
incentive is not conveyed or merely expressed without any reasons
or explanations (Brief emotional incentive).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
Table 4.6 Coding Examples
M essage Context
Information
Explanation
ABC is looking at revising its cell
phone policy, and could use
examples o f other projects’
policies. Thanks.
Factual
context
General information about
the project without detailed
explanations or
interpretations
W e are anticipating the release o f
the DEF Financial Sector/Pension
Development Project ABC around
the end o f January. W e had initially
considered partnering with PBMI,
but they have decided to Prime. We
are researching other partners.
Factual and
interpretive
context
Detailed explanations about
the background information
o f the project (such as
participant, time) provided
I especially would like your
opinion on m y strategy to locate
potential buyers for the firm’s
sesame seeds and sesame seed oil.
Explication of
MM intent
Intention expressed that the
knowledge seeker was
seeking verification limited
to a focused area
In order to best interpret the
concept o f competitiveness in this
project, any perspectives o f
understanding this concept would
be appreciated.
Explication of
MB intent
Intention conveyed that the
knowledge seeker was open
to various opinions on a
concept
W e ARE very eager to find the
right candidate soon.
Brief
emotional
incentive
Information about feelings
provided without any
explanations or reasons
The due date for our response is
November 15. W e really need to
print this by Friday, November 11,
so we can get it out via DHL Given
the urgency o f this task, we are
eager to receive your prompt
response.
Detailed
emotional
incentive
Information about feelings
provided with detailed
explanations which may
attract attention
4.3.3 Boundary Information
Knowledge exploration in organizations requires technical area boundary
spanning (i.e., expertise) and unit (i.e., department) boundary spanning (Rosenkopf
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
& Nerkar, 2000). Boundary information was obtained through both the coding
process and the survey. In the survey, knowledge seekers provided the information
which department (i.e., region) and practice network they belong. The coding
process obtained the information about which department and practice network that
knowledge holders belong.
In this study, departmental boundary (i.e., department) and expertise
boundary (i.e., practice network) were coded. “0” refers to the situation that both the
knowledge seeker and holder were from the same department and have the same
expertise. “ 1” refers to the situation that the knowledge seeker and holder were either
from different departments or have different expertise. “2” refers to the situation that
the knowledge seeker and holder were from different departments and have different
expertise.
4.4 Data Analytic Techniques
It was expected that achievement o f intended learning (i.e., MB learning,
MM learning) mediates the relationship between argument quality and perceived
usefulness o f the knowledge received. Further, detailed situational context
information, focused cognitive context information, and low-level emotional context
information are expected to link to achievement o f intended model maintenance
learning. In contrast, summarized situational context information, open cognitive
context information, and high-level emotional context information are expected to
link to achievement o f intended model building learning. M oreover, the more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
boundaries to cross in the process o f e-mail-mediated knowledge transfer, the more
significant the influence o f corresponding contextual information on achievement of
intended learning.
Based on the objectives o f this dissertation and the sample size, Partial Least
Squares (PLS) was used as the primary data analysis technique (Chin, 2003). PLS is
a commonly used technique in IS research when scholars’ prim ary objectives are
theory development and testing. PLS is suitable for testing hypotheses when the
theory is under development and the precise relationship between the variables is not
known3. It is also suitable for testing the moderation effect proposed in this study
(Chin, 2003). The sample size was also sufficient for our data analysis method. For
PLS (consistent with the heuristic for multiple regression), the sample size should be
10 times the maximum number o f paths leading into any one construct in the
structural equation model (Tabachnick & Fidell, 1989).
PLS is also useful to simultaneously test for alternate structural models. For
example, in addition to the hypothesized relationships, I also tested for a direct
association between (1) interpretive situational context (also, low explication of
learning intent, low level emotional incentive) and achievement o f intended model
building learning, and (2) factual situational context (also, high explication of
learning intent, high level emotional incentive) and achievement o f intended model
maintenance learning. Insignificant path estimates for these two sets o f relationships
would help to confirm the importance o f specific contextual information for the
3 A detailed bibliography and discussion of PLS is at http://disc-nt.cba.uh.edu/chin/index.html.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
particular type o f learning. The detailed results o f the PLS analysis are reported
the next chapter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
CHAPTER 5 DATA ANALYSIS AND RESULTS
The results indicated that Hypotheses 1, 2, 2a, 3, and 3a (see Table 5.1)
were supported; whereas, hypotheses 4 and 4a were not supported. Hypothesis 5 was
partially supported. In the process o f cross-boundary knowledge seeking,
achievement o f intended learning mediates the relationship between argument
quality and perceived usefulness o f the knowledge received through e-mail
(Hypotheses 1). Further, factual and interpretive context and explication o f MM
intent positively affect the achievement o f intended model maintenance learning
(Hypotheses 2 and 3). Factual context and explication o f MB intent positively affect
the achievement o f intended model building learning (Hypotheses 2a and 3 a).
Furthermore, the relationship between situational context (and explication o f
learning intent) and achievement o f intended learning is moderated by the boundaries
that need to be crossing during the process o f e-mail-mediated knowledge transfer
(Hypothesis 5). The relationship between emotional incentive and achievement o f
intended learning was not supported by the results (Hypotheses 4 and 4a). The
moderation effect o f boundary on the relationship between emotional incentive and
achievement of intended mental model learning was not supported (Hypothesis 5).
The results o f descriptive statistics will be presented first (section 5.1). Next
the results o f tests o f convergent and discriminant validity among constructs will be
discussed (section 5.2). This will be followed by a discussion o f the results o f testing
structural model according to Partial Least Square (PLS) analyses (section 5.3).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.1 Summary o f Hypotheses
89
_______________________________ Hypotheses_______________________________
Hypothesis 1: In the process o f electronic knowledge transfer across
boundaries, achievement o f intended mental m odel learning mediates the
relationship between argument quality and perceived usefulness o f the
knowledge received through e-mail.________________________________________
Hypothesis 2: In the process o f electronic knowledge transfer across
boundaries, factual and interpretive context specified in e-mail by knowledge
seekers positively affects their achievement o f intended m odel maintenance
learning._________________________________________________________________
Hypothesis 2a: In the process o f electronic knowledge transfer across
boundaries, factual context specified in e-mail by knowledge seekers positively
affects their achievement o f intended model building learning._______________
Hypothesis 3: In the process o f electronic knowledge transfer across
boundaries, explication o f MM intent specified in e-mail by knowledge seekers
positively affects their achievement of intended model maintenance learning.
Hypothesis 3 a: In the process o f electronic knowledge transfer across
boundaries, explication o f MB intent specified in e-mail by knowledge seekers
positively affects their achievement of intended model building learning.______
Hypothesis 4: In the process o f electronic knowledge transfer across
boundaries, brief emotional incentive specified in e-mail by knowledge seekers
positively affects their achievement of intended model maintenance learning.
Hypothesis 4a: In the process o f electronic knowledge transfer across
boundaries, detailed emotional incentive specified in e-mail by knowledge
seekers positively affects their achievement o f intended m odel building
learning.________________________________________________________________
Hypothesis 5: During the process o f electronic knowledge transfer, the more
organizational boundaries to be crossing, the more significant the impact o f the
features o f defining advice-seeking request (i.e., specification o f situational
context) on achieving intended mental model learning._______________________
Hypothesis 5: During the process o f electronic knowledge transfer, the more
organizational boundaries to be crossing, the more significant the impact o f the
features o f defining advice-seeking request (i.e., specification o f explication of
learning intent) on achieving intended mental model learning.________________
Hypothesis 5: During the process o f electronic knowledge transfer, the more
organizational boundaries to be crossing, the more significant the impact o f the
features o f defining advice-seeking request (i.e., specification o f emotional
incentive) on achieving intended mental model learning._____________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
5.1 Descriptive Statistics
Table 4.4 contained the actual questions for each construct. Means, ranges,
and standard deviations of each construct are reported in Table 5.2. The data was
subjected to reliability tests (Table 5.4). For multi-item reflective constructs, the
reliability estimates were calculated using Cronbach’s alpha, which was sufficiently
above 0.8, indicating good internal consistency (Fomell & Larcker, 1981).
Cronbach’s alpha is a more conservative reliability measure than composite
reliability that is reported in Table 5.4.
Table 5.2 Descriptive statistics
Construct Sample average
Scale 1-7
Standard
deviations
Argument Quality 3.34 0.71
M odel Building Learning (QI) 3.26 0.96
M odel Maintenance Learning (QI) 3.17 0.93
M odel Building Learning (QII) 3.07 0.79
M odel M aintenance Learning (QII) 2.82 0.93
Achievement o f intended MB 1.92 0.31
Achievement o f intended MM 1.83 0.27
Knowledge Usefulness 3.13 0.84
The data were collected when knowledge seekers sent e-mail solicitations to
diverse e-mail distributions lists. Specifically, practice networks involve eight e-mail
distribution lists focusing on different areas o f specialties: agriculture/agribusiness;
democracy and governance; environment and natural resources; finance and banking;
health; humanitarian response; gender; and private sector development. Each o f these
expertise distribution lists consists o f 10-30 people. In addition, there are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
departmental e-mail distribution lists: Africa, Asia, Europe and Eurasia, Latin
America and Caribbean, and M iddle East.
5.2 Discriminant Validity
The discriminant validity o f constructs was shown in two ways. First, results
o f the PLS component-based analyses were presented in Table 5.3. For discriminant
validity to be supported, items must load higher on their own construct than other
constructs (Chin, 1998). Items that are cross-loaded should be eliminated. Items that
had low loadings (below .50) on their own construct or leaded on another construct
at or above .50 should be considered for elimination.
Table 5.3 PLS component-based analysis: Cross-loadings
Construct/Items Argument
Quality
MB Learning M M Learning Knowledge
Usefulness
ArguQualil 0.739 0.348 0.319 0.413
ArguQuali2 0.803 0.412 0.397 0.427
ArguQuali3 0.711 0.376 0.349 0.351
MB1 0.259 0.897 0.067 0.487
MB2 0.513 0.915 -0.104 0.464
MB3 0.417 0.785 -0.073 0.367
MB4 0.353 0.879 0.002 0.483
MM1 0.472 0.099 0.882 0.373
MM2 0.319 -0.061 0.861 0.385
MM3 0.501 -0.020 0.857 0.367
MM4 0.334 0.034 0.884 0.299
KnowUsel 0.464 0.516 0.378 0.831
KnowUse2 0.386 0.469 0.431 0.789
KnowUse3 0.473 0.387 0.381 0.802
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
Table 5.3 provides the correlations o f each item to its intended construct (i.e.,
loadings) and to all other constructs (i.e., cross loadings). A lthough there are a few
cross-loadings, all items load more highly on their own construct than on other
constructs and all constructs share more variance with their measures than with other
constructs. This demonstrates that the correlations of each construct with their own
items are significantly higher than the correlations between the constructs and items
measuring other constructs. The results supported discriminant validity.
Next, discriminant validity was confirmed in another w ay by examining the
Average Variance Extracted (AVE). The AVE4 is a measure o f the variance that a
construct captures from its items relative to the amount due to measurement error
(Chin, 1998). Discriminant validity is confirmed when the AVE by each construct is
larger than the inter-construct correlations (Chin, 1998). Comparing the square root
o f the AVE (i.e., the diagonal elements in Table 5.4) w ith the correlations among
constructs (i.e., the off-diagonal elements in Table 5.4) indicates that each construct
is more closely related to its measures than to those o f other constructs. In sum,
these results support the convergent and discriminant validity o f the constructs.
The correlations were examined for the independent variables and
dependent variables (Table 5.4). Argument quality is highly correlated with
achievement o f intended MB and achievement o f intended MM. In addition,
perceived knowledge usefulness is highly correlated with achievement o f intended
4 AVE is calculated as follows: ( 2 X ; 2) / ( ( X A ;2) + 2 (1 - A ; 2)). In the equation, A ; represents the factor
loading for a construct’s item.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
MB and achievement o f intended M M. Also, achievement o f intended learning was
collected based on cross-sectional data (i.e., differences between the scores from
Questionnaire I and II). These indicate that multicollinearity m ay exist among the set
of predictors to knowledge usefulness (Hair Jr. et. al., 1998).
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 5.4 Inter-Construct Correlations
Construct (#
of items)
Composite
Reliability
Argu
Quality
Achieve
o f MB
Achieve
o f MM
KnowUse Boundary Richness of
SituContext
Context x
Boundary
Learning
Intent
Intent x
Boundary
Emotional
Incentive
IncentiveX
Boundary
Argument
Quality
0.889 0.751 *
Achievement
o f MB
- 0.429 -
Achievement
of MM
- 0.451 -0.095 -
Knowledge
Usefulness
0.932 0.336 0.387 0.443 0.807 *
Boundary - 0.103 0.009 0.101 0.162 -
Richness of
Situational
Context
0.151 0.219 0.257 0.131 0.023
Context x
Boundary
- 0.026 0.215 0.208 0.101 0.032 0.028 -
Explication
o f Learning
Intent
0.187 0.268 0.291 0.112 0.102 0.010 0.016
Intent x
Boundary
- 0.045 0.219 0.233 0.111 0.047 0.012 0.031 0.112 -
Emotional
Incentive
- 0.163 0.292 0.287 0.116 0.051 0.108 0.027 0.173 0.031 -
Incentive x
Boundary
- 0.109 0.334 0.276 0.042 0.013 0.106 0.073 0.061 0.075 0.172 -
*Diagonal elements are the Average Variance Extracted (AVE) estimated for each construct. Off-diagonal elements are the
correlations among the construct
VO
95
5.3 Structural Model
The results o f testing the proposed m odel using PLS are discussed in section
5.3.1. Further, the results o f testing the alternate structural paths are discussed briefly
in section 5.3.2.
5.3.1 Testing Proposed Model
Figure 5.1 and 5.2 present graphical depictions o f the PLS results o f the
overall model (including hypotheses 1 through 5). In PLS, the structural model is
assessed by R-squared values o f the endogenous constructs and should be interpreted
as in multiple regression analysis. From Figure 5.1, most o f the paths are significant
with the model accounting for 54 percent o f the variance in achievement o f intended
model maintenance learning and 46 percent o f the variance in perceived usefulness
o f knowledge received.
The mediation effect o f achievement o f intended learning on the
relationship between argument quality and perceived knowledge usefulness is
supported. Argument quality significantly (with path coefficient 0.495) contributes to
achievement o f intended model maintenance learning. Achievement of intended
model maintenance learning significantly (with path coefficient 0.431) contributes to
perceived knowledge usefulness. However, argument quality did not significantly
(with path coefficient 0.131) contribute to perceived knowledge usefulness without
the mediator. This confirmed the mediation effect o f achievement o f intended model
maintenance learning. The similar findings were obtained when testing the mediation
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
effect of achievement o f intended model building learning. Taken together, these
results confirmed the mediation effect of achievement o f intended learning on the
relationship between argument quality and perceived knowledge usefulness.
Figure 5.1 Structural M odel (associated w ith Model M aintenance Learning)
* Significant at .05 level; ** significant at .01 level
The results in Figure 5.1 support the direct relationship between interpretive
context and achievement o f intended model maintenance learning. Further, boundary
is a moderator in this relationship. Likewise, the relationship between low
explication o f learning intent and achievement o f intended model maintenance
learning is supported, and boundary moderates the relationship. However, the effect
o f low-level emotional incentive on achievement o f intended model maintenance
learning is not supported. Additional tests were conducted to verify that there was no
0.131
A chievem ent o f
intended M M
Learning
' A
Argument 0.495**
Quality
Explication o f MM
intent
Brief
Emotional
> R2 =0.538
0.431**
P erceived
U sefulness o f
K now ledge
received
R2 =0.462
Boundary
Factual and '
Interpretive context
0.286*
incentive
0.128
Factual and Interpretive
context X Boundary
\
Explication of MM intent X
Boundary
\
Brief Emotional incentive
X Boundary
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
multicollinearity effect arising from the interaction terms so as to render emotional
incentive as not significant in the full model (Figure 5.1). The moderation effect o f
boundary is not significant in the relationship between low-level emotional incentive
and achievement o f intended model maintenance learning.
Figure 5.2 Structural Model (associated with M odel Building Learning)
*Significant at .05 level; **significant at .01 level
According to Figure 5.2, most o f the paths are significant with the model
accounting for 55 percent o f the variance in achievement o f intended model building
learning and 51 percent o f the variance in perceived knowledge usefulness. The
direct relationship between factual context and achievement o f intended model
building learning is supported. Further, boundary is a moderator in this relationship.
Similarly, the relationship between high explication o f learning intent and
0.147
r
A chievem ent o f
intended M B
Learning
'A
Argum ent 0.529**
Quality
Explication of
MB intent
Detailed
Emotional
R2 =0.553
0.483**
Perceived
U sefulness o f
K now ledge
received
R2 =0.510
Boundary
Factual context
0.412* 0.478** incentive 0.153
Factual context X Boundary
\
Explication o f MB intent
X Boundary
Detailed Emotional
incentive X Boundary
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
achievement o f intended model building learning is supported, and boundary
moderates the relationship. However, the effect o f high-level emotional incentive on
achievement o f intended model building learning is not supported. Again, additional
test were conducted and there was no multicollinearity effect. The moderation effect
o f boundary is not significant in the relationship between high-level emotional
incentive and achievement o f intended model building learning.
5.3.2 Testing Alternate Structural Paths
In order to further confirm the proposed research model, the relationship
between factual context (and high explication o f learning intent, high level emotional
incentive) and achievement o f intended model maintenance learning is tested. Also,
the relationship between interpretive context (and low explication o f learning intent,
low level emotional incentive) and achievement o f intended model building learning
is tested. The path estimate between factual context (and explication o f MB intent,
detailed emotional incentive) and achievement o f intended model maintenance
learning is 0.057 (0. 079, 0.048 respectively). The path estimate between factual and
interpretive context (and explication o f M M intent, brief emotional incentive) and
achievement o f intended model building learning is 0.069 (0. 051, 0.083
respectively). These paths are not found to be significant. These results further
confirmed the proposed structural model. Discussions o f the results are provided in
the next chapter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
CHAPTER 6 DISCUSSION
This chapter summarizes the results (section 6.1), discusses the limitations of
this dissertation study (section 6.2), highlights the contributions to the current
literature (section 6.3), and presents ideas for future research (section 6.4) as well as
contributions to practice (section 6.5). The chapter ends with a summary o f the
research findings and conclusions (section 6.6).
6.1 Discussion o f the Results
The hypotheses developed in Chapter 3 were tested and results are
summarized in Table 6.1. The measurement model was validated from the results of
the PLS analysis. That is, the discriminant and convergent validity, as assessed using
the PLS tests, support the construct conceptualization and measurement items.
The findings from this research suggest that achieving intended learning is
critical in predicting perceived usefulness o f the knowledge received through
electronic knowledge transfer processes in online communities. Specifically,
achievement o f intended learning mediates the relationship between argument
quality and perceived usefulness o f the knowledge received through electronic
solicitations. Furthermore, knowledge seekers who seek advice through e-mail
distribution lists across boundaries are more likely to achieve intended mental model
development when they include particular features o f advice-seeking request in their
electronic solicitations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
Table 6.1 Summary o f the Results
Hypotheses Results
Hypothesis 1: In the process of electronic knowledge transfer across
boundaries, achievement of intended mental model learning mediates the
relationship between argument quality and perceived usefulness of the
knowledge received through e-mail.
Supported
Hypothesis 2: In the process of electronic knowledge transfer across
boundaries, factual and interpretive context specified in e-mail by
knowledge seekers positively affects their achievement of intended
model maintenance learning.
Supported
Hypothesis 2a: In the process of electronic knowledge transfer across
boundaries, factual context specified in e-mail by knowledge seekers
positively affects their achievement of intended model building learning.
Supported
Hypothesis 3: In the process of electronic knowledge transfer across
boundaries, explication of MM intent specified in e-mail by knowledge
seekers positively affects their achievement of intended model
maintenance learning.
Supported
Hypothesis 3a: In the process of electronic knowledge transfer across
boundaries, explication of MB intent specified in e-mail by knowledge
seekers positively affects their achievement of intended model building
learning.
Supported
Hypothesis 4: In the process of electronic knowledge transfer across
boundaries, brief emotional incentive specified in e-mail by knowledge
seekers positively affects their achievement of intended model
maintenance learning.
Not
Supported
Hypothesis 4a: In the process of electronic knowledge transfer across
boundaries, detailed emotional incentive specified in e-mail by
knowledge seekers positively affects their achievement of intended
model building learning.
Not
Supported
Hypothesis 5: During the process of electronic knowledge transfer, the
more organizational boundaries to be crossing, the more significant the
impact of the features of defining advice-seeking request (i.e.,
specification of situational context) on achieving intended mental model
learning.
Supported
Hypothesis 5: During the process of electronic knowledge transfer, the
more organizational boundaries to be crossing, the more significant the
impact of the features of defining advice-seeking request (i.e.,
specification of explication of learning intent) on achieving intended
mental model learning.
Supported
Hypothesis 5: During the process of electronic knowledge transfer, the
more organizational boundaries to be crossing, the more significant the
impact of the features of defining advice-seeking request (i.e.,
specification of emotional incentive) on achieving intended mental
1 model learning.
Not
Supported
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
Regarding emotional incentive, additional analyses were conducted to verify
its effect. First, correlations were (i.e., (1) emotional incentive and situational context
information; (2) emotional incentive and explication of learning intent; (3) emotional
incentive and the interaction o f situational context information and explication of
learning intent calculated to check the multicollinearity problem. The results showed
that none o f these correlations was significant. Second, emotional incentive
(high-level and low-level) was regressed directly on achievement o f intended
learning (model building and model maintenance, respectively) but its effect is not
significant. Third, in order to further find out if emotional incentive might influence
achievement o f intended learning under different conditions, the relationship
between use o f emotional incentive and demographic variables was studied. Taylor
(1996) stated that demographic variables may have an effect on individual
information behavior, and m ay help us understand behavior o f a restricted population
(e.g., gender different). Hence, a T-test was conducted to investigate the relationship
between emotional incentive and gender. It is plausible that female knowledge
seekers are more likely to use emotional incentive in their e-mail requests than
males, and further, have emotional incentive influence their m ental model learning.
The results (n=109, F=0.69, p=0.41) indicated that there was not a difference in the
relationship between emotional incentive and achievement o f intended learning for
female versus male. All these results indicated that emotional incentive does not
significantly influence achievement o f intended mental model learning or
cross-boundary knowledge access in the e-mail settings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
6.2 Limitations
As w ith any research, this dissertation study has some limitations. First, the
data collection took place at a single site (section 6.2.1). Second, the data were
collected from the perspectives o f knowledge seekers only (section 6.2.2). In
addition, cross-sectional data were used to assess achievement o f intended learning
(section 6.2.3).
6.2.1 A Single Site for D ata Collection
The study concentrated on one single organization and must be extended to
multiple organizations. A single site was used to control for the effect of
organizational differences in electronic knowledge transfer processes. However, a
single site m ay limit the generalizability o f the research findings. For example, most
Company A employees interviewed and surveyed for this dissertation study have
advanced degrees. In addition, findings from the qualitative studies indicate that
most employees agreed that they had a very healthy knowledge sharing culture at
Company A, the communication across departments was open, and e-mail was
common and easy to use. Under these circumstances, deployment o f the features o f a
well-defined electronic solicitation may be more likely to contribute to achievement
o f intended mental model learning. Although the generalizability o f this study may
have been limited, the findings are instructive. This is especially true considering the
continued challenge Company A face getting employees to seek, share, or otherwise
transfer each other’s knowledge through electronic distribution lists across
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
organizational boundaries (e.g. Dougherty, 1992), and the recent calls for the IS
research community to examine alternative uses o f technology for successful
knowledge transfer (e.g. Alavi & Leidner, 2001; Markus, 2001). Moreover, although
the organization (i.e., company A) operates within governmental regulation, it was
similar to any other organization that is independent and conducted itself as a profit
center. Finally, the internationally-operating nature o f the organization w ith a great
deal of ongoing cross-boundary knowledge access makes the findings instructive for
understanding knowledge transfer crossing organizational boundaries.
6.2.2 Perspectives from Knowledge Seekers
All the quantitative data were collected from knowledge seekers, and
knowledge holders’ perspectives m aybe examined to understand a more complete
knowledge transfer process. As we know, there are not only knowledge seekers, but
also knowledge holders in a knowledge transfer process. It is valuable to take
knowledge holders into account to examine the effect o f contextual information on
improving their understanding o f a problem outlined in the request. T e’eni et al.
(2001) studied both the knowledge seeker’s view and the knowledge holder’s
viewpoint in written and recorded communication. However, this dissertation
research focuses on investigating the very early step in an electronic knowledge
transfer process - knowledge seeking. In particular, the focus is on how to define an
advice-seeking request to facilitate achieving knowledge seekers’ intended learning
so as to judge the knowledge received as useful. Thus, following Sussman and Siegal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
(2003), I studied this phenomenon from the perspectives o f knowledge seekers. The
findings from this dissertation study provide a preliminary understanding o f the
effect o f advice-seeking requests sent across boundaries in online communities.
In-depth understanding o f this phenomenon including knowledge holders’ opinions
would open up avenues for future research.
6.2.3 Cross-sectional Data Used
The third limitation is the use o f cross-sectional data to assess achievement
o f intended metal model learning. Intended learning data were collected with
Questionnaire I and achieved learning was assessed with Questionnaire II. The
difference o f the scores between Questionnaire I and II reflected achievement of
intended learning. As such, the mediation effect o f achievement learning may result
from the multicollinearity problem. However, if both intended learning and achieved
learning were collected with the same questionnaire, several things can occur to bias
the results. For example, when respondents are filling out surveys that measure
intended learning and achieved learning simultaneously, 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
may be provided a context that allows them to create and use implicit theories during
survey completion (Podsakoff et al., 2003). Separate collection o f intended learning
and achieved learning was deployed to address these limitations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
6.3 Contributions to Theory
Despite the limitation o f this dissertation study, the study has several
theoretical implications. First, the dissertation research is an initial study addressing
boundary barriers in knowledge transfer through electronic distribution lists in online
communities. M uch o f the IS studies on knowledge transfer tend to focus on other
factors: knowledge workers’ credibility, knowledge w orkers’ expertise, knowledge
w orkers’ involvement, medium involved in knowledge transfer, and characteristics
o f knowledge (e.g., Faraj & Sproull, 1998; Markus, 1994; Nonaka, 1991; Sussman &
Siegal, 2003). Recently, Sussman and Siegal (2003) develop an information adoption
theory to explain advice seeking in the e-mail settings. Their proposed model is an
initial IS study examining knowledge usefulness in e-mail-mediated knowledge
transfer processes. The key findings in Sussman and Siegal’s (2003) study include
that argument quality is critical in predicting usefulness o f the knowledge received
through e-mail. However, theses theories ignore whether the advice comes from the
same or a different thought world (e.g., Sussman & Siegal, 2003) in online
communities, thus assuming that advice across thought worlds is fundamentally the
same as seeking advice from the same thought world. Drawing from the theory o f
thought worlds (Dougherty, 1992) and the theory o f knowledge reuse (Markus,
2001), this study modifies the information adoption model by suggesting the
importance o f bridging thought world barriers in cross-boundary knowledge access
in the electronic settings (i.e., electronic distribution lists) and how to overcome the
barriers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
This study is also insightful by considering the recent call for research to
investigate if intranets can support effective knowledge sharing among specialists
working in a distributed environment (Hollingshead et. al., 2002). According to Fulk
et. al. (2001), intranets can support (1) individual activities, (2) formal information
dissemination, (3) pointers to knowledge and knowledge holders, (4) individual and
group data, information and knowledge sharing, and (5) group interaction via
synchronous or asynchronous methods. Electronic distribution list is a particular type
o f Intranet support to facilitate knowledge sharing across organizational boundaries
in online communities. The findings o f this study suggest feasible mechanisms for
knowledge workers to institute cognitive repairs to naturally flawed human
reasoning.
For example, it is found that, when crossing thought worlds, knowledge
seekers have different learning intentions, and that the effect o f argument quality on
perceived knowledge usefulness is mediated by whether the provided information
meets the knowledge seekers’ learning intentions. The findings suggest that
achievement of intended learning should be incorporated in Sussman and Siegal’s
information adoption theory. Thought world barriers make it hard for knowledge
holders to understand the request sent through e-mail. Thus, the advice contributed
might not be able to m eet knowledge seekers’ needs and will then be judged useless.
Only when knowledge seekers’ needs are met (i.e., achieving intended learning), can
the knowledge received be perceived as useful. While engaged in knowledge work,
knowledge workers must incorporate the knowledge they receive into their schemas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
and mental models, and obtain meaningful learning (Vandenbosch & Higgins, 1996).
W hile quality o f an argument may be an important factor affecting knowledge
usefulness, if the argument quality is not relevant to a knowledge seeker’s intended
learning, argument quality will have minimal useful impact. The finding from this
study adds to the current literature about electronic knowledge transfer (e.g.,
Sussman & Siegal, 2003) by stressing the significance o f m atching intended learning
and achieved learning.
Second, the findings from this dissertation study also contribute to the
thought world literature. When accessing knowledge through e-mail distribution
lists, it is more likely that knowledge workers need to transcend their own thought
worlds to bridge the barriers created by different specialties or interested across
communities. Despite the notion o f interpretive barriers associated with knowledge
transfer across boundaries, Dougherty (1992) did not examine in depth how to
reconcile the barriers. Literature on knowledge reuse (Markus, 2001) was drawn
upon to make suggestions - defining advice-seeking requests w ith critical features:
situational context, explication o f learning intent, and emotional incentive. The
findings from this study further suggest these features m ay facilitate knowledge
seekers achieving their intended learning, and then induce useful knowledge transfer.
Understanding how to overcome thought world barriers contributes to the thought
world literature by extending the theory toward a more complete model.
Third, the findings from this dissertation study contribute to the discussion
about the value o f technology in knowledge transfer. Technologies have been touted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
as a means to w idely disseminate diverse knowledge across organizational
boundaries (e.g., Alavi & Leidner, 2001; Huber, 1990). However, much o f the
research to date has shown that using technologies under such circumstances is
problematic (Alavi & Leidner, 2001; Markus, 2001). The findings suggest that the
failure o f knowledge access across boundaries is in part due to inappropriate
strategies employed in using technology. Dixon (2000) has suggested that the
benefits associated w ith knowledge management technology are limited to the
exchange o f knowledge among people with shared or similar interests. In contrast,
the findings from this dissertation study suggest that proper employment o f strategies
(i.e., use of appropriate features o f advice-seeking requests in electronic solicitations)
could help bridge the barriers to cross-boundary knowledge access through
electronic distribution lists (i.e., from people without shared specialty or interests).
More specifically, this dissertation study suggests that advice-seeking request
features can bridge the barriers to cross-departmental knowledge access in online
communities by providing information about situational context and explication of
learning intent in electronic solicitations.
6.4 Future Research
This dissertation study is fruitful itself also in that it yields several interesting
ideas for future research. For example, this dissertation focuses on examining the
impacts of the choice o f learning made (model building vs. model maintenance) in
e-mail-mediated knowledge transfer cross-boundaries, but not specifically on
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
predicting this choice. It will be interesting to examine w hy knowledge workers
would choose one type o f learning over the other. This area o f research plays a still
unappreciated role in knowledge transfer.
It would be fruitful to examine the views o f knowledge seekers and holders
simultaneously in a similar study. This dissertation study found the use o f proper
dimensions o f advice-seeking requests can positively im pact the perceived
usefulness o f cross-boundary knowledge. It would be interesting future research to
determine to what extent knowledge holders benefit from the different dimensions of
advice-seeking requests provided by knowledge seekers. This m ay further confirm
that different dimensions o f advice-seeking requests influences knowledge transfer
differently.
Future research should also investigate in-depth the impact o f specifying
emotional incentive in technology-mediated knowledge transfer. Conventional
wisdom says that emotional incentive (manifested as em otional information)
influences information processing in physically-shared environments (Forgas, 1991,
2000). Nevertheless, the findings from this dissertation study suggest that emotional
incentive included in electronic solicitations does not significantly affect achieving
intended learning. There are several reasons for this result. It is possible that in the
process o f technology-mediated knowledge transfer, emotional incentive does not
play a role as important as in physically-shared environments due to reduced social
presence. It is also possible that knowledge seekers are not adept to provide
emotional incentive when they send electronic solicitations; and consequently, it is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
hard to find the effect. Future research should do well to conduct additional studies
to clarify these issues.
6.5 Contributions to Practice
This study suggests practical implications as well. W hen knowledge
workers engage in technology-mediated knowledge transfer across boundaries, it is
essential for knowledge seekers to achieve their learning objectives (i.e., intended
learning) in order to perceive the knowledge received as useful. Such objectives
and/or expectations should be communicated and conveyed to knowledge holders
since boundaries create thought world barriers for holders to identify learning needs.
In addition, because the early stages o f using electronic distribution lists are
critical in gaining collective resources necessary for receiving knowledge that will be
perceived as useful, organizations should focus on efforts in the above areas from the
very inception o f distributed work groups (Hollingshead et al., 2002). Specifically, it
is very important for knowledge seekers to deploy strategies (such as properly
specifying situational context and explication o f learning intent) to facilitate the very
early stage o f knowledge transfer processes - defining a knowledge-seeking request.
In online communities, knowledge seekers may or may not know who will read their
questions given the queries are sent to a large electronic distribution list. Knowledge
seekers should not assume that knowledge holders by nature bridge the thought
world barriers and understand their learning needs and underlying situations o f the
problem that they look for help. Therefore, managers need to train the potential
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
knowledge seekers to properly phrase their electronic solicitations. In particular,
when there are more boundaries to be crossing, it will be more important to
strategically define an advice-seeking request in electronic solicitations. As such,
knowledge workers m ay save time and effort in e-mail exchanges (i.e., to identify
what advice would be useful). Further, knowledge holders m ay have resources and
incentives to contribute. It will lead to more effective cross-boundary knowledge
transfer processes. Proper use o f the features o f a well-defined request in electronic
solicitations will finally lead to improvement o f em ployees’ day-to-day work
performance. The results o f the study provide a simple and feasible w ay for
employees to better access cross-boundary knowledge through e-mail interactions
and increase their efficiency.
The findings from this dissertation study also suggest several specific
guidelines for more efficient e-mail usage in organizations, and a better practical
understanding o f knowledge transfer across boundaries in the e-mail settings. For
instance, in order to achieve model building learning (in which mental models are
changed to accommodate new information), employees who send out electronic
inquiries should articulate their situational context (i.e., factual situational context)
and high explication o f learning intent. In contrast, in order to achieve model
maintenance learning (in which new information fits into existing mental models and
confirms them), employees should articulate interpretive situational context and their
explication o f learning intent (i.e., not open to all available alternatives/perspectives)
in their electronic solicitations. These guidelines could be included in a Best
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
Practices document or training program for Practice Network facilitators and
Company A employees. The company would also benefit from a better use of
electronic distribution lists to foster its knowledge-transfer culture in CMC.
6.6 Conclusions
This dissertation study extends the IS literature’s exam ination of knowledge
transfer processes through electronic distribution lists by studying the role of
advice-seeking requests in electronic learning across organizational boundaries.
More specifically, a model proposed that advice provided in e-mail responses would
only be perceived as useful when knowledge seekers achieve intended mental model
learning. In situations where intended learning is not achieved, argument quality may
be o f minimal useful impacts and the knowledge received through e-mail responses
would be not considered useful. In addition, well-defined advice-seeking requests in
electronic solicitations would be positively associated w ith individual’s achievement
o f intended mental model learning. Moreover, this relationship is moderated by
various boundaries that need to be crossing in the knowledge transfer process.
Results from a qualitative and quantitative study supported m ost o f the hypotheses.
Specifically, the findings indicated that achievement o f intended learning mediates
the relationship between argument quality and perceived usefulness o f knowledge
received in electronic knowledge transfer processes. Also, specific features of
advice-seeking requests (i.e., situational context, explication o f learning intent, and
emotional incentives) contribute to achievement o f particular intended mental model
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
learning. Furthermore, the more knowledge exploration boundaries to cross, the
more important it is to deploy appropriate features o f advice-seeking requests in
electronic solicitations. The results from this dissertation study hold promise for
further examination o f related phenomena. Additional research can help build an
in-depth understanding in the area.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
REFERENCES
Alavi, M., and Leidner, D. E. (2001). Review: Knowledge management and
knowledge management systems: Conceptual foundations and research issues. MIS
quarterly, 25(1), 107-136.
Argote, L. (1999). Organizational Learning: Creating, Retaining and Transferring
Knowledge. Kluwer, Norwell. MA.
Argote, L., and Ingram, P. (2000). Knowledge transfer: A basis for competitive
advantage in firms. Organizational Behavior and Human Decision Processes, 82(1),
150-169.
Augier, M., Shariq, S. Z., and Vendelo, M. T. (2001). Understanding Context: Its
Emergence, Transformation and Role in Tacit Knowledge Sharing. Journal o f
Knowledge Management, 5, 125-136.
Bailey, J. E., and Pearson, S. W. (1983). Development o f a tool for measuring and
analyzing computer user satisfaction. Management Science, 29(5), 530-545.
Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology.
Cambridge University Press. London.
Bechky, B. A. (2003). Sharing meaning across occupational communities: The
transformation o f understanding on a production floor. Organization Science, 14(3),
312-330.
Boland, R. J. (1993). Accounting and the interpretive act. Accounting, Organizations
and Society, 18(2/3), 125-146.
Boland, R. J., Tenkasi, R. V., and Te’eni, D. (1994). Designing information
technology to support distributed cognition. Organization Science, 5(3), 456-475.
Brown, J. S., and Duguid, P. (1991). Organizational learning and communities of
practice: Toward a unified view of working, learning, and innovation. Organization
Science, 2, 40-57.
Chin, W. (1998). Issues and opinion on structural equation modeling. MIS Quarterly,
22( 1).
Chin, W. (2003). “PLS Graph”, Soft Modeling Inc.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
Clore, G. L., Schwarz, N., and Conway, M. (1994). Affective causes and
consequences o f social information processing. In R. S. W yer & T. K. Srull (Eds.).
Handbook o f Social Cognition (2nd ed.). Hillsdale, NJ: Erlbaum.
Cossette, P. (1998). The study o f language in organizations: A symbolic
interactionist stance. Human Relations, 15(11), 1355-1377.
Craik, K. (1943). The Nature o f Explanation, Cambridge University Press.
Cambridge.
Cross, R., Rice, R. and Parker, A. (2001). Information seeking in social context:
Structural influences and receipt o f information benefits. IEEE Transactions on
Systems Management and Cybernetics - Part C: Applications and Review, 31(4),
438-448.
Curtis, B., Krasner, H., and Iscoe, N. (1988). A field study o f the software design
process for large systems. Communications o f the ACM, 31(11), 1268-1287.
Daft, R. L., and Lengel, R. H. (1986). Organizational information requirements,
media richness and structural design. Management Science, 32(5), 554-571.
Damasio, A. R. (1994). D escartes’ Error. New York: Grosset/Putnam.
Dennis, A. R., and Kinney, S. T. (1998). Testing media richness theory in the new
media: The effects o f cues, feedback, and task equivocality. Information Systems
Research, 9(3), 256-274.
Dixon, N. (2002). The Neglected Receiver o f Knowledge Sharing. Ivey Business
Journal, 66(4), 35-40.
Dougherty, D. (1992). Interpretive barriers to successful product innovation in large
firms. Organization Science, 3, 179-202.
Faraj, S. and Sproull, L. (2000). Coordinating expertise in software development
teams”, Management Science, 46(12), 1554-1568.
Fiedler, K. (1991). On the task, the measures and the mood in research on affect and
social cognition. In J. P. Forgas (Ed.). Emotion and Social Judgments (pp. 83-104).
Oxford: Pergamon Press.
Forgas, J. P. (1991). Mood effects on partner choice: Role o f affect in social
decisions. Journal o f Personality and Social Psychology, 61, 708-720.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
Forgas, J. P. (2000). Affect and information processing strategies: An interactive
relationship. In J. P. Forgas (Ed.) Feeling and Thinking: The Role o f Affect in Social
Cognition, (pp.253-280).Cambridge: Cambridge University Press.
Fomell, C., and Larcker, D. F. (1981). Structural equation models w ith unobservable
variables and measurement error: Algebra and statistics. Journal o f Marketing
Research, 18(3), 382-388.
Fulk, J. (1993). Social construction o f communication technology. Academy
Management Journal, 36, 921-950.
Fulk, J., Heino, R., Flanagin, A., Monge, P., Kim, K., Bar, F., and Lin, W. (2001).
Intranet functionality as collective action. Unpublished manuscript.
Girin, J. (1990). Problems du language dans les organizsations. In J.-F. Chanlat
(Ed.). L’individu dans 1’organisation: Les dimensions oubliees. Collection “Sciences
de l’administration.” (pp. 37-77). Quebec/Paris: Les Presses de 1 ’ Universite
Laval/Editions ESKA.
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). Prospering in Dynamically Competitive Environments:
Organization Capability as Knowledge Integration. Organization Science, 7(4),
375-387.
Hair Jr., J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate
Data Analysis. Upper Saddle River, NJ: Prentice Hall.
Head, A. J. (2000). Demystifying intranet design: 5 guidelines for building usable
sites. Online, 24, 36-42.
Hollingshead, A., Fulk, J. & Monge, P. (2002). Applying the theories o f transactive
memory and public goods to intranets and knowledge management in organizations.
In P. Hinds & S. Kiesler (Eds.), Distributed work: New research on working across
distance using technology (pp. 335-355). Cambridge, MA: M IT press.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
Hovland, C. I., Janis, I. L., and Kelley, H. H. (1953). Communication and
Persuasion, Yale University Press. New Haven. CT.
Huber, G. P. (1991). A theory o f the effects on advanced inform ation technologies
on organizational design, intelligence, and decision making. Academy o f
Management Review, 75(1), 47-71.
Kettinger, W. J., and Grover, V. (1997). The use o f computer-mediated
communication in an inter-organizational context. Decision Sciences, 28(3),
513-555.
Kirsch, L. J., and Beath, C. M. (1996). The Enactments and Consequences of Token,
Shared, and Compliant Participation in Information Systems Development”,
Accounting, Management and Information Technology, 6(4), 221-254.
Kogut, B., and Zander, U. (1992). Knowledge o f the firm, combinative capabilities,
and the replication o f technology. Organization Science, 3(3), 383-397.
Kraut, R. E ., Fussell, S. R., Brennan, S. E., and Siegel, J. (2001). Understanding
Effects o f Proximity on Collaboration: Implications for Technologies to Support
Remote Collaborative W ork,” In P. Hinds & S. Kiesler (Eds.) Distributed Work.
Cambridge, MA: M IT Press.
Lave, J., and Wenger, E. (1991). Situated Learning: Legitimate Peripheral
Participation, Cambridge University Press, Cambridge, U.K.
Lee, A. S. (1994). Electronic M ail as a Medium for Rich Communication: An
Empirical Investigation Using Hermeneutic Interpretation. MIS Quarterly, 18(2),
143-158.
Levitt, M., and Mahowald. R. (2001). Email usage and forecast analysis. IDC Report
#W25335.
Markus, M. L. (1994). Electronic Mail as the M edium o f M anagerial Choice.
Organization Sciences, 5(4), 502-527.
Markus, M. L. (2001). Toward a theory o f knowledge reuse: Types o f knowledge
reuse situations and factors in reuse success. Journal o f Management Information
Systems, 75(1), 57-93.
Nadler, J., Thompson, L., and Van Boven, L. (2003). Learning negotiation skills:
Four models of knowledge creation and transfer. Management Science, 49(4),
529-540.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
Nonaka, I. (1991). The knowledge-creating company. Harvard Business Review. 69,
96-104.
Nonaka, I., & Takeuchi, H. (1995). The Knowledge Creating Company. New York:
Oxford University Press.
Nonaka, I., Toyama, R. and Nagata, A. (2000). A firm as a knowledge creating entity:
A new perspective on the theory o f the firm. Industrial and Corporate Change, 9(1),
1- 20 .
Norman, D. A. (1982). Learning and Memory, W. H. Freeman, San Francisco.
Oatley, K., and Jenkins, J. M. (1996). Understanding Emotions, Oxford: Blackwell.
O ’Donnell, A. M., and J. O ’Kelly. (1994). Learning from peers: Beyond the rhetoric
o f positive results. Educational Psychology Review, 6(4), 321-349.
Orlikowski, W. J., Yates, J., and Okamura, K. (1995). Shaping electronic
communication: The metastructuring o f technology in the context o f use.
Organization Science, 6(4), 423-444.
Petty, R. E., and Cacioppo, J. T. (1986). Communication and persuasion: Central
and peripheral routes to attitude change, Springer-Verlag. New York.
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: A critical review o f the literature and
recommended remedies. Journal o f Applied Psychology, 88(5), 879-903.
Polanyi, M. (1967). The Tacit Dimension, Garden City. NJ: Doubleday.
Ramirez, A., Walther, J. B., Burgoon, J. K., and Sunnafrank, M. (2002).
Information-seeking strategies, uncertainty, and computer-mediated communication:
Toward a conceptual model. Human Communication Research, 28(2), 213-228.
Rice, R. E. (1992).Task analyzability, use o f new media, and effectiveness: A
multi-site exploration o f media richness. Organization Science, 3(4), 475-500.
Rosenkopf, L. andN erkar, A. (2001). Beyond Local Search: Boundary-Spanning,
Exploration, and Impact in the Optical Disk Industry. Strategic Management Journal,
22, 287-306.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
119
Schwarz, D. (1990). Feeling as Information: Informational and M otivational
Functions o f Affective States”, In. Handbook o f Motivation and Cognition:
Foundation o f Social Behavior, 2 (pp. 527-561), R. M. Sorrentino (Ed.), New York:
Guilford Press.
Smith, K. K. (1984). Rabbits, Lynuxes, and organizational transitions. In J. R.
Kimberly and R. E. Quinn (Eds.). Managing Organizational Transitions (pp.
267-294), Homewood. IL: Irwin.
Sproull, L., and Kiesler, S. (1986). Reducing social context cues: Electronic mail in
organizational communication. Management Science, 32(11), 1492-1512.
Stamm, K., and Dube, R. (1994), The relationship of attitudinal components to trust
in media. Communication Research, 21 (1), 105-123.
Stuart, T. E. and Podolny, J. M. (1996). Local Search and the Evolution o f
Technological Capabilities. Strategic Management Journal, 17, 17-38.
Sussman, W. S. and Siegal, W. S. (2003). Informational Influence in Organizations:
An Integrated Approach to Knowledge Adoption. Information System Research,
14(1), 47-65.
Szulanski, G. (1996). Exploring internal stickiness: Impediments to the transfer of
best practice with the firm. Strategic Management Journal, 17(10), 27-43.
Taylor, R. S. (1996). Information Use Environments, In Ethel A uster and Chun Wei
Choo (Eds.) Managing Information fo r the Competitive Edge (pp.93-135), New York:
Neal-Schuman Publishers.
Tabachnick, B. G., and Fidell, L. S. (1989). Using M ultivariate Statistics, (2nd ed.)
Harper, NY.
Te’eni, D. (2001). Review: A Cognitive-Affective Model o f Organizational
Communication for Designing IT. MIS Quarterly, 25(2), 251-312.
Te’eni, D., Sagie, A., Schwartz, D. G.„ Zaidman, N. and Amichai-Hamburger, Y.
(2001). The Process o f Organizational Communication: A M odel and Field Study.
IEEE Transactions on Professional Communication, 44(1), 6-20.
Tolman, E. C. (1959). Principles o f purposive behaviourism In S. Koch (Ed.),
Psychology: A Study o f a Science, 2, McGraw-Hill. New York.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
Vandenbosch, B., and Higgins, C. (1996). Information Acquisition and M ental
Models: An Investigation into the Relationship Between Behavior and Learning”,
Information System Research, 7(2), 198-214.
Wastell, D. G. (1999). Learning dysfunctions in information systems development:
Overcoming the social defenses with transitional objects. MIS Quarterly, 23(4),
581-600.
Webb, N. M., and A. S. Palincsar. (1996). Group Processes in the Classroom. In D.
Berliner and R. Calfee (Eds), Handbook o f Research in Educational Psychology
(pp.841-873), London: Prentice Hall.
Weick, K. E. (1991).The nontraditional quality o f organizational learning.
Organization Science, 2, 116-124.
Willoughby, T., Wood, E., McDermott, C., and McLaren, J. (2000). Enhancing
learning through strategy instruction and group interaction: Is active generation of
elaborations critical?. Applied Cognitive Psychology, 14, 19-30.
Wu, C., and Shaffer, D. R. (1987). Susceptibility to persuasive appeals as a function
o f source credibility and prior experience w ith the attitude object. Journal o f
Personality and Social Psychology, 52(4), 677-688.
Zack, M. H. (1993). Interactivity and communication mode choice in ongoing
management groups. Information Systems Research, 4(3), 207-239.
Zack, M. H., and McKenney, J. L. (1995). Social context and interaction in ongoing
computer-supported management groups. Organization Science, 6(4), 394-422.
Zhong, J. J., and Majchrzak, A. (2004). An Exploration o f Cognitive Elaboration on
Learning in Information Systems Development Projects. Information Technology
and Management, 5(1/2), 143-160.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
121
APPENDIX A
Introduction Letter*
Over the next week, we will add Jenny Zhong to all practice network email groups.
Jenny Zhong is a doctoral student at USC who will be conducting research on how to
use email effectively when soliciting assistance from a group o f people on an email
list. To that end she will be observing email traffic on the network email lists
through November.
What it means to you: If she sees you seeking assistance from your colleagues on
the PN, she m ay ask you if you got what you needed and to complete a short (10
minute) survey that will help us understand what should go into an email to increase
its efficacy in securing assistance.
Why: Jenny’s research is directly related to our efforts to improve knowledge
management at Chemonics. As you know, the practice networks are important
knowledge management enablers. They bring together people who share a passion
for a particular topic so that they can share information and help each other
learn. Jenny’s research w ill help us improve our knowledge sharing by email, and
consequently strengthen our practice networks, by telling us, exactly, how to
improve the quality o f email advice seeking, which, in turn, w ill result in better
knowledge sharing by email.
Some background: At the end o f 2003 the SVPs agreed to support a knowledge
management study conducted at Chemonics. (See SYP notes from 12/10/2003.) At
the beginning o f 2004, Jenny Zhong came to Chemonics to prepare a survey for this
research, which explores the relationship between good knowledge sharing behavior
and team performance at Chemonics. The study, designed by three researchers from
USC, Rutgers and UMD, requires a survey be completed by a lot o f our field
staff. We are still understanding how best to execute this research. W hile Jenny was
here she learned her way around Chemonics and our practice networks and
developed the idea for her own dissertation, which is to determine what
characteristics o f an email solicitation for assistance (“advice seeking”) result in the
correct assistance being rendered.
Please contact Amanda Jefferson with any questions or concerns.
* It was sent out by Company A ’s Knowledge M anagement Division prior to data
collection.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
122
APPENDIX B
Questionnaire I
Dear XX,
I’m Jenny Zhong. I am a member o f the Home Office and Practice Network e-mail
distribution list. I am also a PhD student at the University o f Southern California
working with Company A on understanding how Company A workers can get the
best benefit from e-mail advice-seeking. As part o f m y research I have been looking
for messages that seek advice regarding a non-administrative action from others via
e-mail. For your reference, please find attached the e-mail that was sent out to all
Company A employees by your KM division about this research opportunity.
I recently observed the message (Subject: XXX) you posted to XX Practice
Network. As part o f m y research, I would like to ask a few questions about it. It will
take you less than 5 minutes to answer these questions. As academic researchers, we
hold the confidentiality o f your responses in the highest regard. A summary o f the
responses provided by Company A employees will be reported to you.
Please read over this request you posted, and answer the following questions (14
questions) about it. Please hit “Reply” and send me your response. Your timely
response will be meaningful!
Original M essage-----
From: XXX
Sent: XXX
To: XX Region; XX Practice Area
Subject: XXX
Please type in your name (first and last):
We are interested in finding out what you intend to learn from your e-mail
request. Sometimes you intend to learn new perspectives, confirm current
perspective, or change perspectives, on such issues as project scope, business
concepts, uses o f technology, etc. By "perspective", we m ean the w ay you see things
about the problem.
Please simply write in the number representing from l="none at all" to 7="a great
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
extent" that best indicates the extent to which you intended each o f the following by
sending the e-mail request.
1= none at all; 2 = very minor; 3 = some extent; 4 = moderate extent; 5 = substantial
extent; 6 = major extent; 7 = great extent.
1. Develop new ideas:
2. Maintain your perspective on an issue:
3. Reinforce your current beliefs:
4. Gain new perspectives:
5. Cause you to change your mind:
6. Verify your assumptions:
7. Gain new ways to do things:
8. Justify your decisions:
Now, please assess your own expertise and involvement in the subject matter o f the
request. Please enter a value from 1 to 7 at the end o f each question, representing
from 1 = not at all; to 4 = neutral; to 7 = a great extent.
9. How informed are you on the subject matter of this issue?
10. To what extent are you an expert on the subject matter of this e-mail
request?
11. How familiar are you with the subject matter of this issue?
12. How involved are you in the subject matter of this email?
13. How much has the issue discussed in this email been on your mind lately?
14: How much has the issue been discussed by you and your colleagues lately?
In a week or so, I will be asking you about the responses that you will receive. Please
e-mail me if you have any questions.
Your participation is greatly appreciated!
Jenny Zhong
PhD Candidate
Marshall School o f Business
University o f Southern California
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
APPENDIX C
Questionnaire II
Dear XX,
This is Jenny again. I have sent you a survey via e-mail a week ago about the
message (Subject: XXX) you posted. Thank you so much for taking time to
complete the survey and send it back to me. Hopefully you have received the
responses that you need.
In order to understand the outcome o f your request via e-mail, please pick the most
useful response that you have received and answer the following questions. If you
did not receive very useful responses, please pick the most recent response that you
have received. Again, as academic researchers, we hold the confidentiality o f your
responses in the highest regard. No one from Company A w ill ever have access to
your answers. Your participation will be greatly appreciated!
Please copy and paste the response (that you have selected) here:
Please think about the response you identified and answer the following questions.
1. The information in this e-mail is ____ . Please enter a value from 1 to 7
representing from 1 = valuable; to 7 = worthless;
2. The information in this e-mail is ____ . Please enter a value from 1 to 7
representing from 1 = informative; to 7 = uninformative;
3. The information in this e-mail is ____ . Please enter a value from 1 to 7
representing from 1 = helpful; to 7 = harmful;
4. The information in this e-mail is . Please enter a value from 1 to 7
representing from 1 = relevant; to 7 = irrelevant;
5. The information in this _e-mail is ____ . Please enter a value from 1 to 7
representing from 1 = complete; to 7 = incomplete;
6. The information in this _e-mail is ____ . Please enter a value from 1 to 7
representing from 1 = consistent; to 7 = inconsistent;
7. The information in this E-mail is . Please enter a value from 1 to 7
representing from 1 = accurate; to 7 = inaccurate;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
8. How knowledgeable is the person who wrote this message, on the topic of the
message:
Please enter a value from 1 to 7 representing from 1 = N ot knowledgeable; to 7 =
Knowledgeable;
9. To what extend is the person who wrote this message an expert on the
message topic:
Please enter a value from 1 to 7 representing from 1 = N ot expert; to 7 = Expert;
Please tell us, overall, the extent to which the responding e-mail message led you
to...
1= none at all; 2 = very minor; 3 = some extent; 4 = m oderate extent; 5 = substantial
extent; 6 = major extent; 7 = great extent.
10. Maintain your perspective on an issue:
11. Reinforce your current beliefs:
12. Develop new ideas:
13. Gain new perspectives:
14. Justify your decisions:
15. Gain new ways to do things:
16. Cause you to change your mind:
17. Verify your assumptions:
Finally, please tell us more about yourself (for statistical purposes):
Which region are you working in? ______
Please list the Practice Networks that you are participating:______
What is your gender? 1. m ale 2. fem ale______
What is your age? <25 yrs ___ 25-35 yrs 36-50 y rs 5 0 + yrs
How long have you worked on this project? ___ years ___
months
How long have you worked for Company A? ___ years months
What is the country o f your national origin? ____________
Please check the category P h.D .:________ MS + specialized courses:
that best represents your jM S:_________ BS + specialized courses:
level of formal education: B S : High School + specialized courses:__
Thank you very much for your participation! Your timely response is meaningful!
Sincerely,
Jenny Zhong
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
PhD Candidate
Marshall School o f Business
University o f Southern California
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Information systems capabilities and indirect relational value: Sustaining networked organizations
PDF
Augmenting knowledge reuse using collaborative filtering systems
PDF
Influencing individual innovation through technology features that support cross -departmental understanding
PDF
IT -enabled dynamic capabilities in new product development: Building a competitive advantage in turbulent environments
PDF
Information diffusion and the boundary of market efficiency
PDF
Does adding nonfinancial value drivers to a summary financial measure improve the learning and performance of managers?
PDF
Inventory and pricing models for perishable products
PDF
An empirical investigation of everyday low price (EDLP) strategy in electronic markets
PDF
Customer satisfaction, customer bargaining power, and financial performance
PDF
Effects of redundancy in media coverage on nonprofessional investors' earnings forecasts
PDF
An investigation of repeat visit behavior on the Internet: Models and applications
PDF
Beyond skill development: The effects of training and development on the attitudes and retention of employees
PDF
Comparison of variance estimators in case -cohort studies
PDF
Biopharmaceutical strategic alliances: Interorganizational dynamics and factors influencing FDA regulatory outcomes
PDF
"How will I feel about it?": The asymmetric impact of affective misforecasting on consumer satisfaction
PDF
Essays on organizational forms and performance in California hospitals
PDF
Assessment of prognostic comorbidity in hospital outcomes research: Is there a role for outpatient pharmacy data?
PDF
A socio-technical approach to support collaborative engineering design
PDF
Foreign capital inflows and trade policy change: The automotive and consumer electronics industries in Brazil and China
PDF
Developing business communication skills: Leveraging stage versus global processes of change in skills improvement approaches
Asset Metadata
Creator
Zhong, Jiangfan Jenny
(author)
Core Title
Defining the advice -seeking request in electronic solicitations in boundary spanning environments
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
), El Sawy, Omar A. (
committee member
), Fulk, Janet (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-463211
Unique identifier
UC11336557
Identifier
3196926.pdf (filename),usctheses-c16-463211 (legacy record id)
Legacy Identifier
3196926.pdf
Dmrecord
463211
Document Type
Dissertation
Rights
Zhong, Jiangfan Jenny
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
Tags
business administration, management