A research study concerning knowledge systems use and the relationships between perceived usefulness, intent to use and level of use

Material Information

A research study concerning knowledge systems use and the relationships between perceived usefulness, intent to use and level of use
Morse, Richard Allen
Publication Date:
Physical Description:
xvi, 204 leaves : illustrations ; 28 cm


Subjects / Keywords:
Organizational learning ( lcsh )
Information resources management ( lcsh )
Knowledge management ( lcsh )
Information resources management ( fast )
Knowledge management ( fast )
Organizational learning ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 193-204).
General Note:
School of Education and Human Development
Statement of Responsibility:
by Richard Allen Morse.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
42618195 ( OCLC )
LD1190.E3 1999d .M67 ( lcc )

Full Text
Richard Allen Morse
B.S., University of Nebraska at Omaha, 1974
M.B.A., Kearney State College, 1981
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Educational Leadership and Innovation

1999 by Richard Allen Morse
All rights reserved.

This thesis for the Doctor of Philosophy
degree by
Richard Allen Morse
has been approved

Morse, Richard Allen (Ph.D., Educational Leadership and Innovation)
A Research Study Concerning Knowledge Systems Use and the Relationships
between Perceived Usefulness, Intent to use and Level of use
Thesis directed by Associate Professor Brent G. Wilson
My purpose in conducting this study is twofold. First, I assembled
literature from knowledge management, organizational learning, and the use of
technology to show how knowledge systems can support actionable knowledge
construction. Knowledge systems are a web-based computer technology used to
access organizational knowledge stored in a central knowledge base. Second, I
gathered empirical data on relationships between two technology acceptance
variables: (1) perceived usefulness and (2) intent to use and Level of
Technology Use, a proposed model framed from merging three existing
technology usage models.
Specifically, I seek to determine if employee perceptions of a knowledge
systems usefulness influence their intent to use the system and how extensively
they have adopted the technology. Knowledge system use is defined by how
completely employees have integrated that technology into their work and task
Questionnaires were distributed to 615 employees of Sequent Computer
Systems, using Sequents email system. Two hundred and three (203)
employees responded to the questionnaire. Data collection involved capturing
both quantitative data from the closed-ended questions and qualitative data from

the open-ended questions. Data analysis included Chi-square measures of
association and oneway ANOVA statistical tests.
Results from data analysis reported statistically significant relationships
between the three technology acceptance variables: perceived usefulness,
intention to use, and level of technology use. Additionally, job category
recorded a statistically significant relationship with perceived usefulness, and
average weekly knowledge system use recorded statistically significant
relationships with the three technology acceptance variables. Respondents
reported several additional factors that contributed to their level of technology
use, including the quality, access, and organization of online information.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Brent G. Wilson

My thanks to my advisor, Brent Wilson, for his patience with me during
my dissertation research. My additional thanks to Laura Goodwin for her
assistance with research methods and statistical analyses. I also wish to thank
Vickie Wood for her assistance in conducting the inter-rater reliability tests.
Finally, I send many sincere thanks to Vickie for her words of encouragement
and hours of camaraderie.

1. INTRODUCTION.............................................1
General Problem.......................................3
Rationale for Study...................................7
Theoretical Framework.................................9
2. LITERATURE REVIEW.......................................15
Technology Adoption................................. 16
Levels of Technology Use and Implementation..........24
Technology Acceptance................................28
Perceived Usefulness...........................29
Intention to Use...............................31
Demographics....................................... 31
Age.......................................... 33
Computer Experience............................33

Knowledge Management.................................... 34
Knowledge Management Model........................35
Knowledge Acquisition.............................36
Knowledge Creation................................37
Knowledge Distribution............................38
Knowledge Storage and Retrieval...................38
Knowledge transfer and utilization................39
Organizational Learning..................................40
Individual Learning...............................42
Systems thinking................................. 43
Managing Mental Models............................44
Quest for Personal Mastery........................46
Team Learning.....................................46
Shared Vision.....................................47
Technology and Knowledge Systems.........................48
Technology Subsystem..............................49
Computer Supported Collaborative Work.............53
Knowledge Systems.................................55
Supported Processes...............................56
Web-Based Knowledge System Structure..............59

3. METHOD......................................................65
Research Questions.......................................65
Study Design.............................................66
Rationale for Sample Size..........................69
Rights of Human Subjects.................................71
Demographic Variables..............................72
Level of Technology Use............................72
Perceived Usefulness.............................. 73
Intention to Use................................. 74
Contributing F actors..............................74
Instrumentation......................................... 75
Procedures............................................. 78
Data Analysis......,.....................................79
4. RESULTS.....................................................83
Level of Technology Use..................................84
Perceived Usefulness....................................,86
Intention to Use.........................................91

Contributing Factors......................:..............94
Open Ended Questions......................................97
Technology Acceptance Variables..........................115
Perceived Usefulness, and Intention to Use........116
Demographic Variables.............................121
5. ANALYSIS....................................................155
Relationships between Variables..........................156
Age and Gender....................................156
Computer Experience...............................157
Job Category......................................158
SCEL Experience...................................160
Frequency of SCEL Use.......................... .160
Contributing Factors............................. 161
Technology Acceptance.............................164
Level of Technology Use...........................166
Technology and the Management of Knowledge...............168
Significant Findings.....................................171
Attributes of a Technology........................175

Directions for Future Research............ 178
B. INVITATION TO PARTICIPATE.......................187
E. EMAIL MESSAGE.................................. 192

2.1 Generic Knowledge Management Model................................35
2.2 Technology Subsystem..............................................51
2.3 Processes supported by Knowledge Systems.........................57
2.4 Web-Based Knowledge System Architecture..........................60
5.1 Model of Technology Implementation................................164

3.1 Inferential Statistics Analyses.....................................80
4.1 Frequency Table Level of Technology Use............................84
4.2 Frequency Table Perceived Usefulness..............................88
4.3 Means and Standard Deviations Perceived Usefulness................89
4.4 Item Analysis Perceived Usefulness................................90
4.5 Frequency Table Intention to Use..................................92
4.6 Means and Standard Deviations Intention to Use....................93
4.7 Item Analysis Intention to Use....................................94
4.8 Frequency Table Contributing Factors..............................95
4.9 Means and standard deviations Contributing Factors................96
4.10 Responses to Would/Would not use SCEL...............................98
4.11 Responses to positive factors question.............................101
4.12 Responses to additional negative factors...........................103
4.13 Responses to further comments question.............................105
4.14 Responses to SCEL use then and now................................ 108
4.15 Correlation Perceived usefulness and intent to use............. 116
4.16 Means and standard deviations Perceived usefulness...............117
4.17 ANOVA Perceived usefulness by level of technology use............117

4.18 Homogeneity of Variance Perceived usefulness by intent to use....118
4.19 Means and standard deviations for intention to use.................119
4.20 ANOVA Intention to use by level of technology use................119
4.21 Homogeneity of Variance Intention to use by level of technology use... 119
4.22 Frequency Table Gender, Age, and Job Category....................122
4.23 Frequency Table-Computer and SCEL Experience and SCEL Use..........123
4.24 Statistics for SCEL Experience and SCEL Use........................125
4.25 Crosstabulation gender and level of technology use...............126
4.26 Chi-square statistics gender and level of technology use.........126
4.27 Means and standard deviations for perceived usefulness by gender...127
4.28 ANOVA -Perceived usefulness by gender..............................127
4.29 Homogeneity of variance Perceived usefulness by gender...........128
4.30 Means and standard deviations Intention to use by gender.........128
4.31 ANOVA Intention to use by gender.................................129
4.32 Homogeneity of variance Intention to. use by gender..............129
4.33 Crosstabulation Age and level of technology use..................131
4.34 Chi-Square statistics Age by level of technology use.............132
4.35 Means and standard deviations Perceived usefulness by age........133
4.36 ANOVA Perceived usefulness by age................................133
4.37 Homogeneity of variance Perceived usefulness by age..............133
4.38 Means and standard deviations Intention to use by age............134

4.39 ANOVA Intention to use by age....................................134
4.40 Homogeneity of variance Intention to use by age...................134
4.41 Cross tabulations Job category and level of technology use........137
4.42 Means and standard deviations Perceived usefulness by job category.... 140
4.43 ANOVA Perceived usefulness by job category.......................140
4.44 Homogeneity of variance Perceived usefulness by job category......140
4.45 Means and standard deviations Intent to use by job category......142
4.46 ANOVA -Intention to use by job category............................142
4.47 Homogeneity of variance Intention to use by job category..........142
4.48 Crosstabulations-Computer experience and level of technology use....144
4.49 Chi-square Statistics-Computer experience and level of technology use... 144
4.50 Means and standard deviations-Perceived usefulness by computer
experience......................................................... 145
4.51 ANOVA Perceived usefulness by computer experience.................146
4.52 Homogeneity of variance-Perceived usefulness by computer experience. 146
4.53 Means and standard deviations-Intention to use by computer experience. 147
4.54 ANOVA Intention to use by computer experience.....................147
4.55 Homogeneity of variance Intention to use by computer experience..147
4.56 Correlations........................................................148
4.57 Means and standard deviations-SCEL experience by level of technology

4.58 ANOVA SCEL experience by level of technology use..............149
4.59 Homogeneity of variance-SCEL experience by level of technology use... 150
4.60 Means and standard deviations SCEL use by level of technology use... 151
4.61 ANOVA SCEL use by level of technology use....................152
4.62 Homogeneity of variance SCEL use by level of technology use..152

In their 75th Anniversary issue, the Harvard Business Review (1997)
posed this question to Peter Drucker and Peter Senge, both respected authors in
the field of business: What problems or challenges do you see already taking
shape for business executives as they move into the next century? (p. 18).
Drucker and Senge identified changes that were not technical or rational in
nature as much as cultural: how to lead organizations that create and nurture
knowledge and how to maintain, as individuals and organizations, our ability to
learn. ZubofF(1988) notes that todays organization may have little choice but
to acquire the learning habit, since in a knowledge-based economy one of its
principal purposes becomes the expansion of knowledge. This knowledge is not
essential for its own sake (as in some academic pursuit), but knowledge that
comes to reside at the core of what it means to be productive. Learning is no
longer a separate activity that occurs either before one enters the workplace or in
classroom settings. Learning takes place as a by-product of people doing their
work. The behaviors that define learning and the behaviors that define being
productive are one and the same.

When employees learn, they construct actionable knowledge, knowledge
that easily translates into performance-enhancing behavior. More specifically, it
is organizational knowledge, the information embodied in the set of task-
environment specific work practices, theories, skills, processes, and heuristics
shared by a firms employees (Argyris, 1996a; Argyris & Schon, 1987). An
organization's task environment is a pattern of interconnected roles operating
through a set of norms, strategies, and assumptions which specify how work gets
divided and how the tasks get performed (Dixon, 1994). For employees to be
productive in their task environment, they must access information on how to
best do their jobs, how others have solved similar problems, and how to make
complex decisions using organizational knowledge gathered by merging diverse
information sources.
Central to any organizational learning environment is the effective use of
technology to provide employees information needed to solve problems, make
decisions, and take effective action (Marquard, 1996; Senge, 1990). Knowledge
systems are centralized computer systems where employees store, structure, and
access a corporations document-based knowledge. Knowledge systems take a
large, diverse collection of document-based knowledge, provide a physical
infrastructure for storing those documents, and provide a common, shared
interface for retrieving information. Physically, the document-based knowledge
is stored in relational databases, encapsulated in word processor files,

spreadsheet files, and graphical presentation files and presented to the users
through a ubiquitous Internet browser-like interface.
Knowledge systems support organization learning by providing
employees with a shared interface to access information during problem solving
and decision-making situations and convert this information into performance-
enhancing, actionable knowledge. How employees perceive a knowledge
system's usefulness in providing critical information and how they perceive a
knowledge systems usefulness in their task environment critically influences
the extent to which the system is integrated into the organizational setting.
General Problem
This study seeks to determine if employee perceptions of a knowledge
systems usefulness influence their intent to use the system and how extensively
they have adopted this technology. Knowledge system use is defined by how
completely employees have integrated that technology into their work and task
environments. When employees possess positive attitudes about the
technology's usefulness, they show an increased likelihood to use the technology
and integrate it into their daily work activities.
Organizational investments in computer-based tools to support enhanced
productivity are risky (Davis, Bagozzi, & Warshaw, 1989). Organizations
achieve no return on investment when employees refuse to adopt a knowledge

system because they do not perceive it as enhancing productivity. Unless
employees perceive the knowledge system as useful, they will retrench to deeply
engrained, traditional knowledge sharing practices such as interpersonal
interaction or paper-based delivery.
Davis, Bagozzi, and Warshaw (1989) report that perceived usefulness is
a key affective variable in predicting an individuals intent to use a technology.
Employees perceive a knowledge system as useful if a subjective probability
exists that this specific technology will increase their performance. After
individuals decide to use a technology, they progressively integrate it into their
work environment. A knowledge systems primary purposes are providing a
common interface for the sharing of organizational knowledge and providing a
repository for employee access to information necessary for decision making
and problem solving. As employees progress toward integrating a knowledge
system into their work environment, they become increasingly dependent on it
as their primary information source. They rely less on tacit knowledge sharing
with co-workers and less on retrieving information from books, reference
manuals, and other paper documentation.
In summary, intent to use and perceived usefulness are variables that are
related to an individuals willingness to use a knowledge system. As individuals
develop stronger attitudes concerning a knowledge system's effectiveness, they
progressively integrate the knowledge system into decision making, problem

solving, and work activities until employees become so dependent on it as an
information source, that if the technology is removed, their productivity
diminishes (Reiber & Welliver, 1989).
This study serves a twofold purpose. First, a significant collection of
literature was assembled on knowledge management, organizational learning,
and the use of technology to support employees constructing actionable
knowledge in their workplace. Action learning in an organizational context
refers to accessing, applying, refining, and constructing rules, heuristics,
processes, and best practices for how employees should do their jobs more
effectively (Sequent, 1996a). Organizational learning literature establishes a
social, cultural, and economic framework that perpetuates two core knowledge
management processes: the creation and sharing of actionable knowledge.
Knowledge systems comprise the technology that enables organizational
The amount of literature on knowledge management and organizational
learning is growing. However, these topics appear to be growing
dichotomously. A significant portion of the literature review synthesizes these
topics and integrates literature on knowledge systems as a learning and
performance enabler. To-date no literature attempts to do this.
Second, an empirical study was conducted gathering empirical data on
relationships between perceived usefulness, intent to use, and Level of

Technology Use. Perceived usefulness and intent to use are variables in Davis,
Bagozzi, and Warshaw's (1989) Technology Acceptance Model. Level of
Technology Use, is a proposed model framed from merging three models: Level
of Use of the Innovation (LoU) (Fullan & Promffet, 1977), Level of Technology
Implementation (LoTi) (Moersch, 1995) and the Technology Infusion Model
(Reiber & Welliver, 1989). This proposed model is a significant contribution to
difiusion/adoption and technology acceptance literature. Current research on
technology acceptance and implementation lacks the holistic yet integrated
perspective presented in this study.
Data collected from the empirical study should answer the following
research questions:
1. Which Level of Technology Use describes how extensively knowledge
workers use knowledge systems?
2. Do knowledge workers perceive a knowledge system as useful in doing their
3. Do knowledge workers intend to use a knowledge system whenever possible?
4. What relationships exist between the Level of Technology Use, perceived
usefulness, and intent-to-use?
5. What relationships exist between gender, age, job category, years of
computer experience, experience using a knowledge system, frequency of

knowledge system use and the Level of Technology Use, perceived
usefulness, and intent-to-use?
Rationale for Study
Our fascination with technological innovations stems from its ambiguity
with existing paradigms (Moersch, 1995). Does technology represent things like
computers, multimedia devices, or other hardware peripherals; or processes, like
financial systems, manufacturing systems, or knowledge management systems;
or infrastructure (Norman, 1998) where the computer disappears behind the
scenes and task-specific solutions (e.g. knowledge management) emerge? Each
perspective on technological innovations has unique attributes and leads the
individual to different implementation strategies.
Yet things and processes are inseparable elements of a larger composite
or cluster of technologies (Rogers, 1995). Rogers states that technology clusters
are individual innovations that cannot be adopted separately, where technologies
converge to form inseparable entities (Norman, 1998). For example, if
knowledge workers are to adopt knowledge management processes (Davenport
& Prusak, 1998), they must concurrently adopt the knowledge system. If
knowledge workers adopt the knowledge system, they must also adopt the use of
web browser software, word processing software, spreadsheet software, and
graphical presentation software. Therefore, the researcher must not only

evaluate individual attitudes toward the knowledge system, but also the degree
to which individuals have implemented the knowledge system as part of their
knowledge management processes. Currently only limited research has been
conducted to support the relationship between attitudes concerning a technology
and the extent of its use.
Several models, Level of Use of the Innovation (LoU) (Fullan &
Promfret, 1977), Technology Acceptance Model (TAM) (Davis, Bogazzi, &
Warshaw, 1989), Level of Technology Implementation (LoTi) (Moersch, 1995)
and the Technology Infusion Model (Reiber & Welliver, 1989) were developed
to explain adoption and integration behavior. All models, except LoTi, are
based on research conducted during the 1970s and 1980s using a population
characterized as technology naive. The LoTi model aligns conceptually with
LoU, but as of this date has not been validated against a population.
Today's businesses recognize the need to collaborate, both externally
between business partners and internally between employees and work teams, in
knowledge sharing and information exchange. As firms escalate collaborative
efforts, a need emerges to leverage the greatest power of information and
information technology, its openness. To maintain control over information
and information technology by restricting its use defeats its purpose. Openness,
however, places the impetus on knowledge workers to develop and refine
knowledge seeking strategies using computer technologies. Conger (1998)

asserts that what we now see is that a phenomenal skill gap, regarding computer
literacy, is growing between the generations. Previously wisdom came from
experience and age. Today wisdom is increasingly tied to youth, thanks largely
to very rapid rates of technological change
As our population reaches maturity in implementing technology and
technologies become more pervasive in our lives, new paradigms emerge. A
new generation of technology-sawy knowledge workers, called Generation X
(Conger, 1998), are less interested in understanding how a technology works
than in how the technology can help them solve problems, make better
decisions, and enhance their job performance. They perceive the technology as
infrastructure, a basic service that is required for them to do their work
(Norman, 1998). Thus the variables which predict or measure adoption
behavior, identified in the LoU, TAM, Technology Infusion and LoTi models,
merit further evaluation.
Theoretical Framework
Knowledge systems cannot improve performance if they are not used.
This research seeks to understand how peoples acceptance of knowledge
systems relates to their intentions to use the knowledge system and how their
intentions in turn relates to their attitudes and their perception of the knowledge
systems usefulness.

Understanding why people accept or reject computer technology is one
of the most challenging issues in Information Systems research (Davis, Bagozzi,
& Warshaw, 1989). Additionally, researchers seek to understand why
individuals choose to only implement the technology minimally or progressively
integrate it into their work activities. If employees perceive that a knowledge
system adds no significant benefit to their ability to learn, e.g. to construct
actionable knowledge that subsequently enhances their performance, they either
reject its use or minimize the degree to which the knowledge system is
integrated into their task environment. It is not sufficient for organizations
simply to avail employees of massive computational power and vast data storage
environments. Employees must integrate the technology into task environments
before organizations can realize productivity gains.
Davis, Bagozzi, and Warshaw (1989) tested the Technology Acceptance
Model (TAM), which posits that individuals rely on two variables, perceived
usefulness and ease of use to form a third variable, their behavioral intention to
use. Davis and colleagues posit that a relationship exist between behavior intent
to use and technology adoption behavior. Szajna (1996) reports on two
variables, perceived usefulness and intent-to-use, as measures of a technologys
effectiveness. Szajna provides supportive evidence that beliefs concerning the
usefulness of technology influences actual technology adoption behavior.
Individual intentions to use a knowledge system, either initially or continuously,

are thus linked to employee perceptions of how useful it is in providing them
with information to be transformed into actionable knowledge; knowledge
which aids the individual in decision making, problem solving, and taking
effective action.
However, attitudes represent only a predisposition to act; an attitude does
not guarantee that the actual behavior occurs (Summers, 1970). Technology
adoption models represent usage behavior as a series of levels ranging from non-
use to complete integration into their work environment and ultimately beyond
the work environment.
Fullan and Promfret (1977) developed the Levels of Use (LoU) of the
Innovation, a model that examines innovation adoption from an individual
perspective. Loucks (1977) reports that individuals progress through eight
levels of use during the Innovation-Adoption process. Yet the LoU model is
conceptually "innovation neutral" which suggests that it lacks the power to
explain innovation specific usage behavior.
Computer technology implementation models (Moersch, 1995; Rieber &
Welliver, 1989) assert that individuals progress up usage levels by increasing the
degree to which the technology is integrated into their work environment. These
models suggest that as individuals' self-effectiveness in using the technology
increases, they increase their use of the technology until the technology is so

ingrained into their work activities that if the technology is not available, the
individual's productivity diminishes.
In summary, employee attitudes concerning a knowledge system's
usefulness as an information source influences their intent to use the technology.
However, once knowledge workers begin using the knowledge system, how do
we determine the extent to which they have integrated the knowledge system
into their work environment? The purpose of this study is to examine
relationships between employee perceptions of usefulness, their behavioral
intent to use, and actual usage behavior.
This study employs both quantitative and qualitative research methods.
The quantitative portion uses survey research, one form of non-experimental
research design. Researchers use survey research to collect information on
variables of interest by administering a questionnaire to a sample population. In
the qualitative research segment, interviews were conducted to collect richer
descriptive data often not available from survey instruments. The interviews
served to both confirm survey results and explore themes that emerged from
responses to the open-end questions.
This study's population is employees of a firm that implemented a
knowledge system approximately three years ago. The firm is a high-

technology computer products and services supplier located in Beaverton,
Oregon. The firm currently employs approximately 2,500 employees with
corporate offices in Beaverton, Oregon and field offices nationally and
internationally. A sample is created by systematically selecting from an
alphabetized list of employee names stored an enterprise-wide email directory.
A pilot study was conducted using ten employees from the Denver, CO
office. Results from this pilot study enabled this researcher to initially test and
retest a new instrument. The pilot study also served to determine if logistical
issues existed in administering the questionnaire and collecting the data. A
secondary objective was to determine if participants experienced difficulty in
understanding the questionnaire and forming appropriate responses.
Currently and during this study I am an employee of Sequent Computer
Systems. In addition I am a user of SCEL (Sequent Corporate Electronic
Library). My role as the researcher in this study was that of participant-observer.
This empirical study is grounded in a sound theoretical framework. The study
utilizes established technology acceptance and use models to evaluate
participants responses. Therefore, I introduced no biases into that segment of
the study. However, in evaluating responses to the qualitative section of this
study, my interpretation of those results may have been biased by my
observations of employees using SCEL and remarks from my peers regarding
their experiences in using SCEL. Every effort was made to minimize the effects

of that bias including the use of another doctoral student to conduct inter-rater
agreement tests.

Individuals exhibit a wide range of behaviors when implementing a
technological innovation. Behaviors include acquiring information about the
technology, implementing that technology to enhance performance, and
progressively integrating the technology into their task environment. Prior to
and during implementation, individuals form attitudes about the technologys
usefulness which are causally linked to these technology adoption behaviors. It
is the relationship between perceived usefulness and implementation that forms
the framework for this study.
This section contains a review of literature on technology
implementation, technology acceptance, knowledge management, active
knowledge construction, and knowledge system technologies. These broad and
diverse topics are problematic to understanding the complex relationships
between a technology user, the technology, and the application of that
technology in a work environment.

Technology Adoption
Rogers (1995) states that before organizational members can integrate a
knowledge system into their ongoing work activities, the organization must first
clarify to organizational members, the meaning of the new technology. When a
new technology is implemented by an organization, it is surrounded by
uncertainty. How does it work? What does it do? Who in the organization is
affected by it? How will it affect me? These are some questions that are
clarified when employees socially construct a shared understanding of the
knowledge system's role in their work life and of how the organization expects
them to interact with the knowledge system in their work environment.
Once an innovation's organizational purpose is clarified, the organization
moves toward routinization (Rogers, 1995); when the technology becomes
incorporated into the regular activities of the organization. Individuals no longer
think of the knowledge system as a new idea. It becomes completely absorbed
in the organization's ongoing activities. For example, when new organizational
knowledge becomes available, the publishers place the documents in the
knowledge system database repository, notifying affected employees with an
email message and an imbedded hypertext link pointing to the new document.
Even though an organization decides to implement a knowledge system,
each individual independently chooses to adopt or reject the innovation. Rogers
(1995) refers to this as an optional innovation-adoption decision. Rather than

use the knowledge system to assist in decision making and problem solving,
individuals may continue to use traditional sources of information such as peers,
written documents, or supervisors. Additionally, individuals may refuse to share
information with other organizational members, thus stifling the firms
knowledge economy.
Individuals follow a five-stage process when deciding to adopt or reject
an innovation (Rogers, 1995). Upon confronting an innovation, they begin
gather information about the innovation, information about how the technology
works and how it can be used in their work environment. Based on information
collected in Stage 1, individuals form an opinion about the innovations'
usefulness which is causally linked to their cognitive decision to use it (Davis,
Bagozzi, & Warshaw, 1989). Rogers asserts that employees form their opinions
based on how complex the innovation is to use, how observable are the benefits,
and how much better is this innovation than the alternative.
At this point, individuals decide to adopt or reject the innovation based
on the strength of their attitudes and perceptions from Stage 2. If the decision is
positive, people implement the technology but to varying degrees. Moersch
(1995) suggests that employee self-confidence is a key variable in determining
the extent to which employees adopt a technology. Finally, individuals look to
their peers, supervisors, and often only to themselves to affirm that their
decision was correct.

Rogers' (1995) model is a general-purpose innovation adoption model
and therefore does not explain innovation-specific behaviors. Moore (1991)
extends the Diffusion Adoption model to specifically relate to technology
adoption. The Technology Adoption Life Cycle's (TALC) underlying thesis is
that individuals within a given community absorb a technology in stages
corresponding to the psychological and sociological profile of various segments
within that community. TALC helps us understand how members of these
segments accept technology.
In order for individuals to better understand the complexity of
technological innovations, Moore (1991) introduces two categories of
innovations, continuous innovations which refers to the normal upgrading of
products and discontinuous innovations which demand significant changes by
not only the end user, but by the infrastructure which services the technology.
Our attitude toward a technology becomes significant anytime we are introduced
to new products (discontinuous innovations) that require use to change our
current mode of behavior. Moore's TALC applies to discontinuous innovations
Moore (1991) and Rogers (1995) agree that individuals can be
categorized based on how aggressively they pursue new innovations, either as
innovators, early adopters, early majority, late majority, or laggards. However,
Rogers asserts that innovativeness is temporal, the degree to which an individual

adopts a new innovation more quickly than other organizational members.
Rogers further suggests that no pronounced breaks in the innovativeness
continuum exist between each of the adopter categories.
Although Moore (1991) uses similar adopter categories, he presents a
series of gaps between the categories which technology implemented must
bridge before the next adopter category will begin using the new technology.
These gaps symbolize the disassociation between two adopter groups that is
the difficulty any group has in accepting an innovation if it is presented in the
same way to that group as it was to more progressive adopter groups.
Moore (1991) suggests that innovators are labeled by their peers as
"techies." They pursue new technology products aggressively because
technology is central to their life regardless of the business function it performs.
Innovators, as gatekeepers (Rogers, 1995) perform useful functions because
their endorsement reassures other adopters that the technology works. An
innovator's attention span with an innovation is typically very short. They may
not progressively integrate a new technology into their work environment
because they typically abandon an existing technology in quest for something
Moore (1991) asserts that early adopters are not technologists; they are
visionaries. They are comfortable using new technologies, but they must find a
purpose for the technology first. Early adopters are technology enthusiasts who

nurture fledgling products and help them gain power and acceptability (Norman,
1998). Often they are labeled as the forward thinkers of an organization
constantly searching to solve new and existing problems with technology-based
solutions. Herein lies the crack between innovators and early adopters, when the
technology product cannot be readily translated into a major benefit. The
innovator loves the architecture but the early adopters cannot figure out how to
start using it. This is the crack through which new innovations fall and are
seldom reintroduced again.
Moore (1991) labels early majority adopters as pragmatists. They are
driven by a strong sense of practicality. Like their early adopter counterparts,
they are comfortable with technology. However, before adopting a new
technology, they require well-established references particularly from other
Although pragmatists rely on early adopters to provide a meaningful
application for the new technology, visionaries posses four characteristics
(Moore, 1991) which can alienate pragmatists during their adoption-decision
process. First, visionaries seeming lack respect for the value of colleagues
experience. Early adopters tend to focus on finding new and often novel uses
for the technology, thus ignoring the pragmatists pleas for practicality.

Second, pragmatists believe that visionaries take a greater interest in a
technology than in their industry by finding solutions that can set new and often
impractical directions for their organization.
Next, when visionaries find an application for a new technology product
they tend not to recognize the importance of the products infrastructure. They
are blinded by the vision, not seeing the secondary and tertiary impacts to
computer networks, business processes, and organizational culture.
Finally, pragmatists perceive visionaries as generally disruptive. The
pragmatists practical nature is comfortable with status quo, therefore disruption
is perceived negatively.
Moore (1991) suggests that the crack between the early adopter and the
early majority categories be likened to a chasm. The early majority adopter
population is the largest group that can be influenced.into using a technology the
earliest in the Technology Adoption Lifecycle. Therefore, Moore considers the
bridge which spans the chasm as the most important challenge facing
technology implementers.
In order for technology implementers to build this bridge they must
understand two key expectations of the pragmatists. First, the early majority
needs to perceive that the technology offers them a productivity improvement.
Visionaries must not only find an application for the technology but also how
that technology can enhance the pragmatist's ability to make better decisions,

solve increasingly complex problems, and take more effective action. Second, a
goal of pragmatists is to make a percent improvement, through incremental,
measurable, and predictable progress. As practical people, they recognize that
magnanimous productivity gains are not possible without complete upheaval, a
state they wish to stay away from. Rogers (1995) suggests when an innovations
benefits are observable, the rate of adoption increases for all adopter categories.
Moore (1991) asserts that this attribute is more important to the early majority
category than the others.
Moore (1991) labels the late majority adopter population as conservative.
They share similar traits with the early majority except that they are not
comfortable with technology. Conservatives are against discontinuous
innovations. They buy and use technology not because they like it but because
they feel they must, just to stay on par with the rest of the world. They adopt a
technology when the product is mature (Norman, 1998). Conservatives are
reluctant to admit to their pragmatist friends that they are unwilling or unable to
step up to the same level of technological self-support. Bridging the crack
between early majority and late majority requires technology implementers to
address the issue of the end users technological competency through training
and mentoring.
Moore (1991) refers to the final adopter category, laggards, as skeptics.
They do not want anything to do with technology. They do not participate in

high-tech discussions except to try to block purchasing decisions. Skeptics,
however, can teach us a lot about what is wrong with a technology. Norman
(1998) cautions us not to give up on laggards. He suggests that even a
technology hater can become part of the late majority if the perceived benefits
are sufficiently great: if the task is important, valuable, and cannot be done in
any other way.
In summary, when a technology is first introduced, the user community
has a large amount of unfilled needs which both the techies and visionaries hope
to satisfy. Moore (1991) and Norman (1998) suggest that when technology
implementers bridge the chasm, they establish a transition point at which the
technology begins fulfilling some basic needs. From this point forward, the
pragmatists, conservatives, and some skeptics view the technology as
infrastructure, a service necessary for the delivery of solutions; where the
technology is masked behind the solution. One important lesson about
infrastructure technologies: it doesnt matter whether or not your technology is
superior; it only matters that what is being offered is good enough to deliver the
solution. Therefore, when organizations implement continuous innovations,
such as a network upgrade or a faster CPU, members of the early majority, late
majority, and laggards categories tend to disapprove of such implementations
because their present infrastructure sufficiently satisfies their needs.

Regardless what form a technology takes, the easy part is changing out
the technology; the difficult aspects are social, organizational, and cultural
(Norman, 1998). Norman states that the digital computer industry is just now
crossing the chasm. The computer has been with us for fifty years, the PC for
twenty years, and the Internet for thirty years. Norman thus concludes, a
technological product can take longer to mature than a person(p.33).
Levels of Technology Use and Implementation
Loucks (1977) reports that individuals transition through eight levels of
use when adopting an innovation. Individuals progress up these levels by
acquiring more information concerning the innovation, increasing their use of
the innovation, sharing information with peers, and assessing how the
innovation impacts their work environment. These eight levels are distinct
states that represent observably different types of usage behavior and patterns of
1. nonuse no action is being taken to use the innovation;
2. orientation the user is seeking out information about the innovation;
3. preparation the user is preparing to use the innovation;
4. mechanical use the user is using the innovation in a poorly
coordinated manner and is making user-oriented changes;

5. routine the user is making few or no changes and has an established
pattern of use;
6. refinement the user is making changes to increase outcomes;
7. integration the user is making deliberate efforts to coordinate with
others in using the innovation;
8. renewal the user is seeking more effective alternatives to
established uses of the innovation.
LoU of the Innovation models behaviors demonstrated by individuals as
they grow in the process of innovation implementation. However, the LoU
model is innovation neutral. When Roecks and Andrews (1980) applied the
LoU model to a technological innovation, they found that their subjects were
unable to differentiate between the eight usage levels. Roecks and Andrews
reported that 35% of the participants were at the Non-Use level and 60% were at
the Routine and Mechanical Use categories.
Reiber and Welliver (1989) developed the Technology Infusion Model, a
model comprised of three levels: familiarization, utilization, and integration.
Individuals progress up these levels by acquiring more information about how to
use the technology and by increasingly integrating the technology into then-
work environment.
At the familiarization level, individuals are not yet using the technology.
They direct their efforts at acquiring the most basic operational information

about the technology, how it works and how it could be used to complete then-
work activities. At the next level, utilization, individuals begin using the
technology. Reiber and Welliver (1989) consider a technology as integrated
when an individuals productivity is diminished if the technology is removed.
Marcinkiewitz (1993; 1994) used the Infusion Model to measure level of use for
teachers using computer technology. He found that teachers usage was
normally distributed across the familiarization, utilization, and integration
Moersch (1995) created a framework conceptually aligned with the
Level of Use of the Innovation Model. The Level of Technology
Implementation (LoTi) framework measures technology implementation
behaviors at seven discrete levels ranging from Non-Use (level 0) to Refinement
(level 6). Moersch suggests that self-effectiveness is the key affective variable
that determines at which level an individual operates. Self-efficacy theory
suggests that individuals with a low level of self-confidence will often choose a
level of innovation use that they can handle.
At the non-use level individuals perceive that they have no time or
access to a technology. They typically use the non-technology equivalent to get
their work done. At level 2, awareness, employees still perceive no relevance to
integrating the technology into their work environment. They might use a peer
to get the information for them. At the next level, exploration, employees begin

using technology to supplement their work activities. Employees still use the
non-technology alternative as their primary solution, but might use the
technology if the alternative cannot provide all of the desired outcomes. At
infusion, level 4, employees begin to augment their work activities with the
technology. Technology is now seen as a tool to enhance their outputs, to make
them better than can be achieved through alternative approaches. In level 5,
employees integrate the technology in a manner that enhances productivity.
Individuals perceive technology as a tool to identify and solve authentic
problems. At the next level, expansion* employees extend technology access
beyond the immediate task environment. Individuals use technology to elicit
applications outside the organization to aid in making decisions, solving
problems, and taking effective action. The final level, refinement, is where
individuals can alter the technology with the objective of bonding the
technology with the workplace. In organizational settings, individuals can rarely
reach this level since changing the technology is beyond their scope of authority
and responsibility.
In summary, literature on level of technology use and implementation
focuses on identifying discrete usage states. Each state describes individual
behaviors on the extent to which employees integrate the technology into their
work processes and is able to enhance their productivity in the process. What is
missing from this literature is prescriptive information on how individuals

progress up the usage levels. Moersch (1995) suggests that an individuals self-
confidence is the key motivational variable. However, Delcourt and Kinze
(1993) report that enhancing self-efficacy can only be achieved by focusing on
granular tasks, rather than an entire technological innovation. No single variable,
either behavioral or affective, operating singularly can affect that type of
change. In the next section, I present literature on models that describe how
individuals accept new technologies and ultimately influence their usage.
Technology Acceptance
Information systems researchers propose intentional models from social
psychology research as a potential theoretical foundation for research on
determinants of user behavior (Davis, Bagozzi, & Warshaw, 1989). Ajzen and
Fishbein (1980) theory of reasoned action (TRA) is a well-researched intention
model that has been successful in examining behavior across a wide variety of
domains and should therefore be appropriate for studying the determinants of
computer usage behavior (Davis, Bagozzi, & Warshaw, 1989).
Davis (1986) adapted TRA in developing the Technology Acceptance
Model (TAM), which specifically examines computer usage behavior. TAM
uses TRA as a theoretical basis for measuring relationships between two key
beliefs: perceived usefulness and perceived ease of use, and three outcomes:
users attitudes, intent-to-use, and actual computer adoption behavior. Davis,

Bagozzi, and Warshaw (1989) tested the Technology Acceptance Model
(TAM), which posits that individuals rely on both their perceptions of usefulness
and ease of use to form their behavioral intention to use which in-tum predicts
adoption behavior.
Szajna (1996) reports that two variables, perceived usefulness and intent-
to-use. measure a technologys acceptance. Szajnas research explains a
relationship between beliefs concerning the usefulness of technology and actual
technology adoption behavior. Individual intentions to use a knowledge system,
either initially or continuously, are thus related to the innovations perceived
usefulness, how it aids the individual in decision making, problem solving, and
taking effective action. Additionally, external variables, not measured in this
study, such as task, user characteristics, political influences, organizational
factors, and the technology implementation process, influence technology
acceptance behavior indirectly by affecting beliefs, attitudes, and intentions
(Szajna, 1996).
Perceived Usefulness
Perceived usefulness is a variable grounded in expectancy theory, a
theory comprised of perceptions that predict human behavior. Within
expectancy theory, there are three component perceptions: (a) valence, the value
or desire one has of some goal; (b) expectancy, the expectation that by one's

effort one is capable of achieving some performance; and (c) instrumentality, the
belief that the achieved performance actually results in the valued goal (Vroom,
For example, expectancy theory suggests that individuals are motivated
to use a knowledge system if they believe that by expending some effort would
result: enhanced productivity, increased knowledge sharing, and increased
organizational learning (expectancy). Additionally, they believe that using the
knowledge system is instrumental in achieving a valued goal (instrumentality)
such as increased knowledge sharing. But only if the goal of increased
knowledge sharing is a goal valued by employees (valence). The perception of
the instrumentality of a behavior is a value strongly associated with the adoption
of that behavior (Marcinkiewicz, 1995; Vroom, 1964).
Perceived usefulness is defined as the prospective users subjective
probability that using a specific technology will increase his or her performance
(Davis, Bagozzi, & Warshaw, 1989). Because this variable is predictive of
behavior, an awareness of the predictive capability of perceived usefulness to
computer use would be valuable information for individuals responsible for
technology implementation.

Intention to Use
Davis, Bagozzi, and Warshaw (1989) assert that a relationship exists
between perceived usefulness and intention to use. This relationship is based on
the ideas that, within organizations, people form intentions toward behaviors
that they believe increase job performance. This is because enhanced
performance is instrumental to achieving extrinsic rewards (Vroom, 1964).
Intentions toward such behavior are based on a cognitive decision to improve
performance. Hence Davis, Bagozzi, and Warshaw hypothesize that the
relationship between perceived usefulness and intention to use in the TAM is a
resulting effect from people forming intentions toward using a new technology
based largely on a cognitive appraisal of how it will improve their performance.
While technology acceptance research has found significant cross-
cultural differences, it has ignored the effects of gender, even though in socio-
linguistic research, gender is a fundamental aspect of culture. Indeed, socio-
linguistic research shows that men tend to focus communication on hierarchy
and independence, while women focus on intimacy and solidarity.
Therefore Gefen & Straub (1997) conclude that discourse is
characterized by patterns of speech that are gender specific. Discourse among

men tends to follow a pattern based on the notion of social hierarchy and focus
on asserting independence and seeking respect. While discourse among women
follows a network oriented pattern and focus on creating intimacy.
In previous research, Wells and Anderson (1997) suggest that before
employees are trained and gain experience on a specific technology, no
significant relationships exist between gender and attitudes concerning
technology use. However after employees used the technology more frequently,
females developed concerns about its relevance to their work environment and
their ability to collaborate with their peers using the technology.
Marcinkiewicz (1995) found no correlation between gender and with
level of computer use and perceived relevance. He attributes these findings to
minimal variation in age (85% between 20-23 years) and gender (92% female).
Gefen and Straub (1997) found that women and men differ in their
attitudes toward email but not in their use. They posit that with a medium used
to exchange ideas and knowledge, individuals vary their degree of use based on
a perceived social presence in the medium. Perceived social presence refers to
the sense of human contact embodied in the medium (Gefen & Straub).
Therefore knowledge workers could perceive a social presence in discourse
patterns embodied in documents stored in an organizations knowledge system.

The relationship of age to computer use is unclear. Studies on the effect
of age on innovation adoption yield mixed results (Marcinkiewicz, 1995).
However, n a research study on educators using computer technology
Marcinkiewicz (1994) reported that no significant correlation exists between age
and level of use.
Conger (1998) indicates that the generation after the 1980s Baby
Boomers are the true information generation, possessing a facility for accessing
and manipulating information. The computer age began with the Baby
Boomers, with the inculcation of the personal computer (PC). Before the PC,
computers were distance, behind the glass walls of data centers. With the arrival
of the PC, computing became truly accessible. Along with their proficiency
with technology, Gen Xers yearn for workplaces that feel like communities.
This propensity to interact with peers coupled with their technology savvy
suggests that this generation is worth studying.
Computer Experience
Experience with a specific behavior influences attitudes toward that
behavior and influences their rate of adopting an innovation (Rogers, 1995).
Experience is also elemental to the technology infusion model, the theoretical
basis for categorizing level of technology use (Reiber & Welliver, 1989).

Marcinkiewicz (1995) reports a significant relationship (p < .001) between
computer experience and perceived relevance and level of computer use.
Delcourt and Kinze (1993) found a significant relationship (p < .005) between
frequency of use and perceive usefulness and self-confidence, which Moersch
(1995) suggests impacts level of technology integration.
Knowledge Management
Knowledge management focuses on understanding how knowledge is
acquired, created, stored and utilized within an organization. Successful
companies are able to acquire, codify, and transfer knowledge more effectively
and with greater speed than the competition (Myers, 1996). Organizations
provide employees with an environment to learn and share knowledge using
technology with the goal of enhancing their productivity.
Learning occurs when individuals create new knowledge by combining
explicit knowledge accessed from knowledge systems, with their prior
knowledge, normally in tacit form. The individual utilizes this new knowledge
to complete his or her task. The individual publishes the resulting new
knowledge into the knowledge system for use by other employees. This cycle of
knowledge creation, publication, and sharing is the central theme of knowledge
management. The following knowledge management model presents four

processes that enable employees, while interacting with their knowledge system,
to generate and share knowledge.
Knowledge Management Model
Knowledge management is a methodology grounded in the generic
process-centric model in Figure 2-1.
Figure 2-1 Generic Knowledge Management Model

Knowledge management processes, discussed individually in the
following sections, Support a firms knowledge economy. Individuals transact
knowledge sharing similar to how they exchange goods and services in a
commodity-based economy. People share products and services for the
betterment of all parties and the collective good of the organization.
Knowledge Acquisition
Organizations acquire knowledge from both external and internal
sources. Methods to acquire information from external sources include:
benchmarking best practices from other organizations; attending conferences;
hiring consultants; monitoring economic, social, and technological trends;
collecting data from customers, competitors, and resources; hiring new staff;
collaborating with other organizations, building alliances, forming joint
ventures, and establishing knowledge links with business partners.
Organizations acquire knowledge internally by tapping into the knowledge of its
staff learning from experience, and implementing continuous process
Two important points exist relative to knowledge acquisition. First,
information, whether it is acquired from an external or an internal source is
subject to perceptual filters (norms, values, and procedures) that influence what
information the organization listens to and ultimately accepts. Second,

knowledge acquisition systemically is guided by a firms core competency
strategy. Individuals search for information, internally and externally, which
enhances performance and adds to existing knowledge bases. For organizations
to meet their strategic objectives, knowledge acquired from multiple sources
must self-organize around the firms key business processes and knowledge
domains modeled in a firms value chain.
Knowledge Creation
Whereas knowledge acquisition is generally adaptive, knowledge
creation is generative, where knowledge is actively constructed from
information previously stored and new information drawn from the environment
(Kozma, 1992). There are a number of activities that an organization can
undertake to create knowledge:
Action learninginvolves working on problems, focusing on the learning
acquired, and actually implementing solutions.
Systematic problem solving--requires a mindset, disciplined in both
reductionism and holistic thinking, attentive to details, and willing to push
beyond the obvious to assess underlying causes.
Learning from past experiences reviews a companys successes and
failures, assessing them systematically, and transferring and recording the

lessons learned in a way that will be of maximum benefit to the
Knowledge Distribution
Core to knowledge management are processes in which multi-
disciplinary knowledge is created and distributed to those who need it. Van der
Spek and Spijkervet (1997) report in a survey of sixty Dutch organizations, that
hardly any structural attention is paid to knowledge management. What is often
lacking is coordination between various activities and departments. Synergy,
necessary to integrate knowledge across multidisciplinary areas, is often
missing. Knowledge distribution processes are charged with disseminating the
best knowledge to the right people in the most cost effective and timely fashion.
Knowledge Storage and Retrieval
In order to store and later to retrieve knowledge, an organization must
first determine what is important to retain and how best to retain it. Knowledge
should be structured and stored so the system can find and deliver it quickly and
correctly. Also the knowledge should be divided into categories such as facts,
policies, and procedures on a learning-needs basis.
When structuring knowledge, it is important to consider how the
information will be retrieved by different groups of people. Functional and
effective knowledge storage systems allow categorization around learning

needs, work objectives, user expertise, use of the knowledge, and location
(where the information is stored). Van der Spek and Spijkervet (1997)
acknowledge two features of knowledge, availability and content, impact its
accessibility. Often, knowledge is not present in its optimal form, is not
available when needed, and is not present where the work activity is carried out.
Additionally, knowledge content is often not complete, not current, and not
Knowledge transfer and utilization
Knowledge transfer and utilization involves the mechanical, electronic,
and interpersonal movement of information and knowledge both intentionally
and unintentionally. Organizations intentionally transfer knowledge by written
communications, training, internal conferences, internal publications, job
rotation and job transfer, and mentoring. Organizations unintentionally transfer
knowledge as a function of unplanned human interaction, i.e. job rotation,
stories and myths, task forces, and informal networks.
The American Productivity and Quality Center, APQC (1996), surveyed
firms who use knowledge management processes. Their studies focus on:
identifying strategies, approaches, and tools for successfully using
information technology to support knowledge management,

leveraging existing technologies and data to build a knowledge system, and
determining current and future requirements for a successful enterprise-wide
knowledge management strategy and architecture.
APQCs research (1997) suggests that firms measure the effectiveness of
a knowledge system using enterprise-wide metrics: innovation, process
improvement, business growth, and customer satisfaction. Current research
does not measure an individuals perception of a knowledge system's usefulness
as an information source and its degree of integration into ongoing work
activities. Although considerable descriptive research exists on knowledge
management and organizational learning, few empirical research studies are
available which measure employee perceptions of and attitudes toward
knowledge systems.
Organizational Learning
Probst and Buchell (1997) define organizational learning as the process
by which the organizations knowledge base changes leading to improved
problem-solving ability and capacity for effective action. Argyris and Schon
(1987) assert that learning occurs when we take effective action, when we detect
and correct errors.
Shank (1997) writes,
If Ive learned anything in my past 15 years working with all types of
companies, both U.S. and international, its that companies learning

systems are bankrupt. The way managers attempt to help their people
acquire knowledge and skills has absolutely nothing to do with the way
people actually learn. Trainers rely on lectures and tests, memorization
and manuals. The problem with training is, it is just like school. They
train people just like the schools teach students: Both rely on telling
and no one remembers much thats taught and whats told doesnt
translate into usable skills (p. ix).
The concept of learning has achieved prominence within management
and educational studies. We should not dismiss it as the latest fad, since the
topic of learning attracts increasing attention both in academic circles and in
business practice. One reason for this heightened awareness is the increasing
pressure of change on companies. The rate of change accelerates steadily and
companies must retain their competitive advantage in an increasingly complex
environment. In the future, learning becomes the only lasting competitive
advantage (deGeus, 1988).
Probst and Buchel (1997) note that most psychological definitions of
learning remain at the level of learning by the individual. System theoreticians
take a different approach: learning by an organization should satisfy the needs of
a collective, focusing attention on the organization as a framework for individual
action. Most analysts who approach organizational learning from this angle give
prominence to interactions between the individual and the organization.
Although the relationship between learning at the organizational level and

learning by individuals is not fully understood, one can say that learning by
individuals is a prerequisite of organizational learning.
Individual Learning
Argyris and Schon (1987) define individual learning as the act of finding
relevant information and applying it to the employees work to make a positive
difference in business results. Manquard (1996) cities Schein (1993) who
defines learning as a process by which individuals gain new knowledge and
insights that result in a change of behavior and actions. Jonassen, Mayes, and
McAleese (1992) suggest that during the process of learning, individuals
construct a reality or at least interpret it based on their mental models. Senge
(1990) contends that learning is ultimately related to action.
Action learning involves working on real problems, focusing on the
learning acquired, and actually implementing solutions (Handy, 1996). Action
learning builds upon the experience and knowledge of an individual or group
and the skilled, fresh questioning that creates new knowledge. Actionable
knowledge is the knowledge of practice (Argyris & Schon, 1987). Knowledge
that is actionable, regardless of its content, contains causal claims. It says, if
you act in such and such a way, the following will likely occur.
People are the pivotal part of managing knowledge. They take data and
transform it into valuable knowledge for personal and organizational use. If

individuals are to acquire the learning habit, they must possess the skills, or
disciplines Senge (1990), of systems thinking, managing mental models, quest
for personal mastery, team learning, and shared vision. Each discipline is
discussed in the following sections.
Systems thinking
Systems thinking is the way we characterize and describe a problem
(Salisbury, 1996). Systems thinkers think of a problem as a system, not just a
linear, cause-to-effect, independent situation. System thinking represents a
conceptual framework one uses to make patterns of relationships clearer, and to
help one see how to change them effectively (Senge, 1990).
Salisbury (1996) suggests that every system has homeostasis (tendency
to repel change). Homeostasis manifests itself as the inclination of a system to
move back to a previous state of equilibrium after being disturbed by external
forces. Organizations attempt to ignore change, and if coerced into changing,
they tend to implement the change while conducting business as usual.
Homeostasis suggests that implementing change is more successful if forces are
intrinsic, thus modifying an organizations systemic structure.
Knowledge systems support systemic change by perpetuating knowledge
creation and knowledge sharing. Organizational systems exhibit a lesser
tendency to repel change if employees feel empowered to develop innovative

solutions and publish those solutions in the knowledge system. Knowledge
sharing diffuses the innovative solutions throughout an organization as
knowledge workers access the newly created knowledge.
System complexity, another system characteristic (Salisbury, 1996), is
managed through feedback systems. Either of two forms of feedback systems,
single-loop and double-loop begins when employees encounter a failure in
expectation (Argyris, 1993). When something doesnt work as expected,
employees either correct the situation and continue with their work activities,
single-loop feedback, or they correct the situation, reflect on how to alter the
procedure, and then continue with their work, double-loop feedback. However,
of significance is that employees manage system complexity through active
learning strategies present in both feedback systems. Knowledge workers use
the appropriate learning strategy, single-loop or double-loop, as part of the
organizational systems tendency to self-regulate. Employees use knowledge
systems to manage system complexity by capturing new knowledge created
during both feedback loops.
Managing Mental Models
Mental models are constructions of how we understand the world, which
then becomes the basis for our actions. For example, our mental models about
learning and work impact how we relate and act on the job. Mental models

significantly impact how knowledge workers create new knowledge. New
insights fail to get put into practice because they conflict with deeply ingrained
assumptions of how the world works, assumptions that limit us to familiar ways
of thinking and acting.
Although people do not always behave congruently with their espoused
theories (what they say, their vision), they do behave congruently with their
theories-in-use or mental models (Arygis, 1996a). Senge (1990) recommends
that organizations implement the practice of managing mental models;
examining deeply ingrained assumptions of reality, sharing those models with
associates and changing the model if appropriate. Unexamined mental models
lead to gaps between the model and current reality. The inertia of deeply
entrenched mental models can overwhelm even the best systemic insights.
People develop defensive routines that insulate their mental models
from examination, and they consequently develop skilled incompetence-an
oxymoron that describes most adult learners who are skilled at protecting
themselves from the pain and threat posed by learning situations (Argyis,
In summary, individuals in learning organizations make decisions based
on shared understandings of interrelationships and patterns of change.
Employees explicate their mental models and exam their peers mental models
by querying organizational knowledge stored in the firms knowledge system.

Managing mental models, the activities of surfacing, testing, and improving
internal pictures of how the world works, promises to be a major breakthrough
for building organizations with robust knowledge management systems.
Quest for Personal Mastery
Personal mastery indicates a high level of proficiency in a subject or skill
area. It requires commitment to lifelong learning so employees can transform
knowledge into expertise and proficiency. Personal mastery requires two
underlying movements before it becomes integrated into our lives: (1)
continually clarifying what is important to us (vision); (2) continually learning
how to see reality more clearly (Senge, 1990). Yet systems thinking suggests
that individuals develop a more holistic, multidisciplinary expertise. Knowledge
systems enable individuals to access multi-disciplinary organizational
knowledge, knowledge that can aid their lifelong learning ventures.
Team Learning
Team learning focuses on the process of aligning, where a group of
people functions as a whole, and developing the teams capacity to learn. Senge
(1990) asserts that team learning requires three critical dimensions:
1. The need to think insightfully about complex problems. Team
members must manage their mental models, allowing new ideas to surface, and
correct their mental models if appropriate.

2. The need for innovative, coordinated actions. Team members must
adopt an operational trust, where each team member remains conscious of other
team members and can be counted on to act in ways that complement others
3. Team members must inculcate the practice and skills of their team into
other teams, thus sharing knowledge and new mental models.
Dialogue is the primary medium for team learning. Dialogue denotes a
high level of listening and communication between people. Through dialogue
individuals learn how to recognize the patterns of interaction in teams that
promote or undermine learning. Dialogue is the critical medium for connecting,
inventing, and coordinating learning and action in the workplace. Knowledge
systems provide a shared interface into a common information space which
Ellis, Gibbs and Rein (1993) contend supports dialogue among groups of people
working on a common task.
Shared Vision
A shared vision is not an idea although it may be inspired by an idea. In
its simplest level, a shared vision is the answer to the question, What do we
want to create? The discipline of shared vision involves the skill of uncovering
shared perspectives of a desired future state that foster genuine commitment
toward a long term objective rather than just compliance (Senge, 1990). Senge

suggests that without shared vision, generative learning, the creation of
something new during the learning activity, cannot occur; it can only occur
when people strive to accomplish something that matters deeply to them.
Knowledge systems provide an interface into a shared information space from
which employees examine shared perspectives.
The topic of learning in an organizational environment attracts attention
from both business and academic communities. Through continuous learning,
organizations adapt and transform their entities into viable, healthy, competitive
forces in their marketplaces. Individuals learn when expectation failures occur;
gaps between espoused theories and theories-in-use. Yet individuals are skillful
at using defensive routines that inhibit learning. Organizations can neutralize
defensive patterns by creating an organizational culture in which learners are
empowered to experiment and where learners are self-directed to enhance their
learning processes.
Technology and Knowledge Systems
A firms technology subsystem does not create knowledge and cannot
guarantee or even promote knowledge generation or knowledge sharing in a
corporate culture that does not favor those activities. The proverbial phrase if

they build it, they will come does not apply to technology (Davenport &
Prusak, 1998, p. 18).
The assumption that technology can replace human knowledge or create
its equivalent has proven false time and again (Davenport & Prusak, 1998).
However, developments in technology are among the positive factors fueling
interest in knowledge management. Networked computing provides new ways
for individuals to share knowledge. Technologies such as Lotus Notes and the
World Wide Web make knowledge easier to collect, store in repositories, and
distribute to desktops. The recent expanding role of intranet use is one
manifestation of the increasing role computer technology plays in
communication and knowledge seeking. However, having more technology will
not improve the state of the information. The computational power of
computers has little relevance to knowledge work, but the communication and
storage capabilities of networked computers make them knowledge enablers.
Businesses are becoming aware both of the potential of this technology to
enhance knowledge work and that the potential can be realized only if they
understand how knowledge is shared and developed.
Technology Subsystem
Marquard (1996) describes the technology subsystem as the supporting,
integrated technological networks and information tools that allow access to and

exchange of information and learning. It includes technical processes, systems,
and structure for collaboration, coaching, coordination, and other knowledge
skills. Marquard lists three major components of the technology subsystem,
shown in Figure 2-2:
1. Information technology (IT) is computer-based technology that gathers,
codes, stores, and transfers information across organizations and across the
2. Technology-based learning utilizes video, audio, and computer-based
multimedia training for the purpose of delivering and sharing knowledge and
3. Electronic performance support systems (EPSS) which use data and
knowledge bases to capture, store, and distribute information throughout the
organization so as to help workers reach their highest level of performance
in the fastest possible time, with the least personnel support.
Each of these components is discussed more fully in the following sections.

Figure 2-2. Technology Subsystem
Information technology. Marquard (1996) suggests that IT can be a key
mechanism for transferring knowledge throughout the organization for the
following reasons:

1. IT can improve the ability of people to communicate with one another
because it blurs the boundaries of the company and increases the range of
possible relationships beyond hierarchies.
2. IT makes it easier for people to communicate directly across time and space.
3. IT reduces the number of management levels needed in the hierarchy, yet at
the same time providing an enhanced potential for span of control.
4. IT contributes to the flexibility with mobile work stations, relational
databases, and the storage of knowledge in open databases rather than in the
minds of individuals.
Clearly information handling which includes storage, movement, and use of
information is the key objective of IT systems.
Technology-based learning. Technology-based learning involves the
utilization of video, audio, and computer-based multimedia training for the
purpose of training and instruction. Technology-based learning is interested in
developing a corporate learning environment which Marquard (1996) suggests
should be employee-controlled. As business dynamics change, employees need
the flexibility to complete their training in their time frames and at their pace.
Electronic Performance Support Systems. Marquard (1996) cites Gloria
Gery, who posits that the goal of Electronic Performance Support Systems
(EPSS), is to provide whatever is necessary to generate performance and
learning at the point of need. On-line help, integrated training and job aids, and

on-line documentation are but a few examples of EPSS systems. However,
EPSS systems traditionally only address a specific performance need, not the
broad issues of organizational learning and knowledge management which are
supported only through knowledge systems. EPSS. can support organizational
learning through performance-centered designs and through generation and
capture of new knowledge (Marquard, 1996).
As organizations struggle to remain competitive, they must adopt new
ways of capturing, organizing, storing, and accessing information from their
internal and external environments. There is also a need to capture collaborative
knowledge for the purpose of providing individuals and the organization with
information to perform current job tasks and adapt to the changing external
environment. The following section offers a framework for Computer
Supported Cooperative Work (CSCW) as a tool for capturing organizational
knowledge and for fostering an environment conducive to learning regardless of
distance or organizational structure.
Computer Supported Collaborative Work
CSCW is the study of the nature of work environments, task
environments, and the technology which supports these environments.
Groupware is one form of CSCW technology which supports groups of people
working on a common task or toward a common goal and that provides an

interface into a shared environment (Ellis, Gibbs, and Rein, 1993). The notion
of common task and shared environment are core elements of groupware and are
significant to the study of knowledge management. Central to the issue of
knowledge management is determining how to increase an individuals
intelligence while working in shared environments.
Additionally, Bannon and Schmidt (1993) suggests that before the work
environment is changed through technological developments, we really need to
understand what goes on in the workplace and the complex interactions between
technology, the work organizations, and the requirements of the task
environment. This is consistent with research by Johnson-Lenz and Johnson-
Lenz (1991) which asserts that groupware contains a human component which is
a shared "mental model" of what the group is doing (purpose), how it is doing it
(process) and the group's values and norms (culture).
CSCW research focuses on the nature of the interplay between
technology, knowledge workers, and their task and work environments. The
objective of a knowledge system is to provide a shared environment and
groupware to increase human intelligence. A knowledge management approach
provides methodologies for creating, embodying, disseminating, and using
intellectual capital. However, knowledge systems must do more than provide an
environment in which technology simply moves data around.

Knowledge Systems
Knowledge systems are centralized computer systems that store, structure,
and provide access to the corporations document-based knowledge.
Knowledge systems take a large, diverse collection of document-based
knowledge, provide a physical infrastructure for storing those documents, and
provide a logical structure for retrieving information. Tobin (1997) identifies
four major components of a knowledge system:
1. A repository, commonly a computerized database, of specific company
knowledge and experience.
2. A directory of the specific knowledge, skills, and experience held by groups
and individual employees throughout the company.
3. A directory of learning resources, within and without the company, that
employees can access to help them plan their own learning activities.
4. A set of tools, methods, and capabilities that enable employees to learn from
each other, and to learn together.
Individuals, regardless of their roles, use a knowledge system with the
objective of enhancing productivity. Therefore a tight integration between core
business practices, which add value to products and services in the firms value
chain, and publication and use processes in the knowledge system is essential if
productivity goals are to be reached. In most cases, knowledge-based systems
support knowledge workers directly by performing knowledge-intensive work.

To do so successfully requires that knowledge-based systems correspond
between the concepts, associations, mental models, conceptual frameworks, and
objectives that the knowledge worker employs to perform his or her work (Wiig,
The rise of Internet technologies and knowledge management disciplines
dovetail nicely in that Internet technologies, and particularly the World Wide
Web (WWW) technologies, are well suited to the creation of intranets that
support key parts of the knowledge management. Intranet web-based
technologies provide a centralized storage, structuring, and dissemination
mechanisms for unstructured and semi-structured data, including multimedia
documents, images, and video.
The following sections present an architectural and functional description of
a knowledge system as implemented at Sequent Computer Systems in
Beaverton, OR. Company white papers, technical manuals, and other company
communications constitute sources for information presented in the following
Supported Processes
A knowledge system supports processes that publish and disseminate
organizational knowledge throughout the firm, along the lines of the process
model described below.

Figure 2-3. Processes supported by Knowledge Systems
Adapted from Sequent, 1996b.
The Document Publication Process. The knowledge system begins its
work with the document publication process, providing the infrastructure for any
publisher to place documents into the organizations knowledge base.

Knowledge systems-supported document publication technology captures not
only the document, but all the metadata on the document required to make the
document accessible to consumers, and manageable by librarians and
management. In addition, the knowledge system uses its own knowledge
concerning which employees are interested in particular classes of information
and which employees are required to review particular classes of information, so
as to notify those employees when new material is created or modified.
Document search, retrieval and enhancement. Knowledge systems also
provide the infrastructure for consumers within the firm to:
search for specific documents, or for general or specific areas of
retrieve any document (subject to security controls) in a form usable by the
consumers local desktop toolkit (wordprocessor, spreadsheet, or email);
enhance any document in the library by attaching more information to the
base document.
Library design and maintenance. Although the core structure of the
knowledge system is frequently built based on a model of the firms internal and
external value chains, many employees still need to view the firm as a set of
discrete functions: manufacturing, finance, and marketing. Others need to see
the firm as a network of named individuals, each of which contributes well-
understood components to the firms knowledge. Still others need to look at the

base of documents in the knowledge system through the lens of the companys
product sets or markets.
The knowledge system also provides a comprehensive set of
mechanisms: to enforce document-level security profiles, to flag and report
events associated with any document, to automatically age documents out of the
library as the information in the documents becomes stale, and to maintain
multiple versions of any one document as documents are edited, modified,
republished and retired.
Web-Based Knowledge System Structure
The basic knowledge system structure is described below.

Figure 2-4. Web-Based Knowledge System Architecture
Adapted from Sequent, 1996b.

Consumers interact with the knowledge system through their Web browser,
which in turn interacts with a Web server that integrates several important
classes of resources:
the content store, which contains the documents published by knowledge
workers, many in multiple formats(word processing or HTML), and multiple
revisions or versions.
the indices, which provide fairly low-level (fiill-text and near-full-text)
access to documents in the knowledge system via searching.
the metadata store, which contains all the information about documents,
publishers and consumers required to manage the document store, provide
appropriate security and access control, construct card catalogs, institute and
manage document aging schemes, and facilitate other administrative and
user access tasks.
external OLTP resources, whose data sets and user interfaces may be
represented in the knowledge system as forms or documents (reports)
external DSS resources, whose data sets and user interfaces may be
represented in the knowledge system as forms or documents (reports).
Consumers also interact with the knowledge system via their mail agent.
The knowledge system uses the messaging system to notify consumers who
have registered their interest in particular topics, knowledge domains or
documents, when new documents are available, when existing documents have

been modified, and when material in the library has been locked or removed
because the information contained in those documents has aged. The publishing
tool and publishing server support the publication process, gather the
appropriate metadata at time of publication and in general facilitate the smooth
and orderly addition of documents to the knowledge system.
Using infrastructural technologies, the World Wide Web and relational
databases and proven design practices, organizations can create centrally-
managed and administered knowledge systems that serve the desktops of the
entire firm, regardless of their location, toolset, or function. In the process
giving the firm its first opportunity to understand fully how knowledge is
constructed, embodied, disseminated and used within the firm.
Technology developments such as the World Wide Web, relational
databases, and sophisticated hypertext search engines and navigational tools
enable knowledge management processes. However, the technology subsystem
simply supports the system-linked processes which define a learning
organization: people, the organizational culture, knowledge, learning, and
knowledge management. Corporate knowledge systems are chartered with
providing a logical and physical structure which allow individuals to access and
create intellectual capital while performing in their work and task environments.

A learning organization is a place where through learning, people
continually reframe their world and their relationships to that world, discovering
how they create their reality and their future. A learning organization adopts a
willingness to identify and challenge its existing paradigms, valuing employee
output and the skills necessary to yield that output, rewarding the thinking not
just the doing, eliciting input and commitment to the vision, values, and
performance expectations, providing opportunities for growth, accepting and
encouraging mistakes through experimentation. It makes use of the learning of
the individual members and encourages and rewards widespread and
spontaneous learning.
The topic of learning in an organizational environment attracts attention
from both business and academic communities. Through continuous learning,
organizations adapt and transform their entities into viable, healthy, competitive
forces in their marketplaces. Individuals learn when expectation failures occur;
gaps between espoused theories and theories-in-use. Yet individuals are skillful
at using defensive routines that inhibit learning. Organizations can mitigate
defensive patterns by creating an organizational culture in which learners are
empowered to experiment, where learners share diverse work experiences, and
where learners are self-directed to enhance their learning processes.

Knowledge management seeks to understand how individuals in
organizational settings acquire, create, store, and use knowledge. Without this
understanding, organizations cannot expect to learn, not only through
knowledge construction, but also through adaptive and transformational learning
Technology developments such as the World Wide Web, relational
databases, and sophisticated hypertext search engines and navigational tools
enable knowledge management processes. However, the technology subsystem
simply supports the system-linked processes that define a learning organization:
people, the organizational culture, knowledge, learning, and knowledge
management. Corporate knowledge systems are chartered with providing both
logical and physical structures that allow individuals to access and create
intellectual capital while performing in their work and task environments.

The purpose in conducting this study is to gather empirical data on
relationships between technology acceptance and level of technology use.
Technology acceptance is measured through two variables: perceived usefulness
and intent to use (Davis, Bagozzi, & Warshaw, 1989). Level of Technology Use
(LOTU) is a proposed model framed from merging three models: Level of Use
of the Innovation (LoU) (Fullan & Promfret, 1977), Level of Technology
Implementation (LoTi) (Moersch, 1995) and the Technology Infusion Model
(Reiber & Welliver, 1989). LOTU measures how extensively individuals
integrate technology into their work activities. Chapter 3 describes the design,
subjects, instrumentation, procedures, variables, and data analysis used in this
research study.
Research Questions
Data collected from the empirical study are intended to answer the
following research questions:

1. Which Level of Technology Use describes how extensively knowledge
workers use knowledge systems?
2. Do knowledge workers perceive a knowledge system as useful in doing their
3. Do knowledge workers intend to use a knowledge system whenever possible?
4. What relationships exist between the Level of Technology Use, perceived
usefulness, and intent-to-use?
5. What relationships exist between gender, age, job category, years of
computer experience, experience using a knowledge system, frequency of
knowledge system use and the Level of Technology Use, perceived
usefulness, and intent-to-use?
Additionally, this research study is intended to provide insight into more
global questions not directly addressed in the previous research questions:
How can technology use in the classroom be changed to better prepare future
knowledge workers?
What reforms are needed if educational institutions are to strive toward the
vision of a learning organization?
Study Design
Research studies designed so that the variables are not directly or
actively manipulated by the researcher are non-experimental designs (McMillan

and Schumacher, 1997). This study uses a survey research methodology, one
form of non-experimental research design. Researchers use survey research to
collect information on variables of interest by administering a questionnaire to a
sample population.
The studys population was employees of Sequent Computer Systems as
of February 1999. Sequent implemented a knowledge system approximately
five years ago. The firm is a high-technology computer products and services
supplier located in Beaverton, Oregon. At the time of this study, Sequent
employed approximately 2,500 employees with corporate offices in Beaverton,
Oregon and field offices nationally and internationally. A sample was
systematically selected from an alphabetized list of employee names in an email
Initially a pilot study was conducted using five employees located in
Beaverton. Based on their feedback, the questionnaire was modified and the
study re-piloted using ten employees located in the Denver, Colorado office.
Participant feedback, gathered from both pilot groups, framed the questionnaire
in Appendix A
This studys population is knowledge workers from Sequent Computer
Systems which markets hardware, software, and services to Fortune 5000

companies. Sample subjects were drawn from employees in both the field
offices, located nationally and internationally, and the corporate offices in
Beaverton, Oregon.
Sample size must be large enough so inferences made concerning the
characteristics of the sample generalize to the target population. A second
concern is whether a probability exists that the characteristics under
investigation in the sample exist representatively in the population. In
probability sampling, subjects are drawn from a larger population in such a way
that the probability of selecting each member of the population is known. This
type of sampling allows researchers to estimate or infer characteristics of a
population from a smaller group of subjects (McMillan and Schumacher, 1997).
Probability samples, while never perfectly representative of the population from
which they are drawn, are typically more representative that other types of
samples (Babbie, 1990). Babbie states, probability sampling remains the most
respected method used by survey researchers today (p. 67).
Probability sampling involves systematically selecting individuals from
the target population. Bias is avoided in systematic sampling because of the
high probability that all of the population s characteristics are represented in the
sample. Systematic sampling is a form of probability sampling in which the
researcher selects every wth subject from a list of all subjects in the population,
beginning at a randomly selected location. The sample for this study was drawn

from a list of employees alphabetized by their first name. McMillan and
Schumacher (1997) suggest using alphabetical lists because they do not create
periodicity, a cyclical pattern found in the list, such as length of employment,
departmental order, geographic sequence, or some other form of categorization,
which prevents selecting a representative sample.
Participants were chosen by selecting the first name and extracting every
fourth name thereafter. All selected email addresses were copied into thirty-two
email aliases, twenty-five addresses per alias. The questionnaire (Appendix A)
and Invitation to Participate (Appendix B) were distributed through company
email one alias at a time until the message was sent to all thirty-two aliases. The
email addresses were partitioned to minimize the load impact on Sequent's email
servers and to expedite the process of maintaining the email aliases.
Rationale for Sample Size.
Krathwohl (1993) suggests that sample size relates to four factors:
certainty of inference desired, precision of inference, homogeneity or
heterogeneity of the population, and effect size.
Certainty of inference desired describes the degree of certainty that the
data collected from the sample characterizes the population. If the researcher
wants to be absolutely sure that the information gathered from the sample
represents the population, a larger sample size is needed. Anderson (1990)

reports that with estimates of certainty and precision, accurate forecasts are
possible from modest samples when the population is large and the sample is
drawn systematically. The sample size for this study was calculated using 95%
confidence and 5% tolerable sampling error (precision of inference).
Precision of inference refers to the degree of preciseness in the data. If
the researcher desires the representativeness of the information to be very
accurate, then a larger sample size is warranted.
The third factor, homogeneity or heterogeneity of the population on the
characteristic of interest refers to the variability of the characteristics under
investigation. When the population is heterogeneous, a larger sample size is
required. Based on observation, I anticipate a homogeneous population,
particularly around the variables of perceived usefulness and intention to use.
The final factor, effect size, is the degree to which the phenomenon is
present in the population as compared to normal sampling variation (Cohen,
1977). Krathwohl (1993) asserts that when normal variability among a
population is large and the effect we are looking for is small by comparison,
normal variability may hide consistent differences.
Effect size can be determined statistically; however the researcher must
know or have a estimate of a populations variability. Without this empirical
data, only reasoned observation is used to measure variability. Based on my

observations, the normal variability among the population should not mask the
variability of the phenomenon under investigation.
Anderson (1990) asserts that for a population of 1,000 the sample size
should be 177; for a population of 5,000 the sample size should be 356. The
firm employs approximately 2,500, therefore by extrapolation, the
recommended sample size is 250.1 selected a sample of 615 allowing for 66%
non-response. Although a 50% response rate is considered adequate for analysis
(Rea & Parker, 1997; Babbie, 1990), of greater interest to the researcher is
whether inferences concerning the variables under investigation can be made
from these responses back to the population.
Rights of Human Subjects
Sequent's Manager of Corporate Knowledge Management granted
permission to conduct the study. A letter granting permission is found in
Appendix C. A letter, in Appendix D, grants use of the firms name in this
document. Participants are asked to cooperate via an Invitation to Participate
found in Appendix B. This letter describes the study, the importance of each
individuals participation, my role, followed by the request for participation and
the guarantee of confidentiality. Participants may sign the questionnaire if they
wish, but are not required to do so. Additionally, the participants were notified
in the email message, Appendix E, of the voluntary nature of their participation.

Demographic Variables
Participants submit information for the following demographic variables:
age, gender, job category, years of computer experience, months of experience
using SCEL, and average weekly SCEL use. I italicized the variables for
purposes of clarity. Gender, and job category are measured on a nominal scale.
Months of experience using SCEL and average weekly SCEL use are measured
using a ratio scale. In this study I used the mean, median, mode, standard
deviation, and frequency tables to describe the characteristics of the interval or
ratio-level data. I used only frequency tables to describe the nominal or ordinal-
level data; gender, job category.
Level of Technology Use
Level of Technology Use is the degree to which organizational members
have integrated a knowledge system into their task environment. Lower
numbered responses suggest little or no use of the knowledge system; higher
numbered responses report degrees of integration of the knowledge system into
employees daily work activities. Response 1 indicates non-use. Response 2
indicates that individuals utilize the knowledge system but not as their primary
choice as an information source. Response 3 or infusion signifies that at this
level individuals are beginning to routinize (Rogers, 1995) the technology.

They now use the knowledge system equally along with alternative information
sources. If respondents select Response 4, they have integrated the knowledge
system into their work environment to the extent that if the knowledge system is
unavailable, their productivity diminishes. Response 5 has the same
characteristics as Response 4 except that employees have now expanded their
use of SCEL to include technology-based information sources outside of then-
task environment.
Each response refers to a ranking or ordering of technology use.
However the interval between each response cannot be characterized as equal.
Therefore, the Level of Technology Use variable is measured on the ordinal
scale. A frequency table was used in conjunction with median and mode
statistics to describe the characteristics of this variable.
Perceived Usefulness
Perceived usefulness measures knowledge workers perceptions of
whether the knowledge system enhances job performance through better
decision making, problem solving, and taking more effective action. The
perceived usefulness construct was measured using nine item-level questions
using a five-point Likert scale, response 1 for strongly agree, response 5 for
strongly disagree, and response 3 as undecided or neutral. The perceived
usefulness item-level variables are measured on the interval scale. Frequency

tables and mean, median, mode, and standard deviation statistics were used to
describe the characteristics of the perceived usefulness variables. Item analysis
(Cronbachs Alpha) and inter-item correlation tests were conducted to measure
relationships between each item-level variable and to estimate the reliability of
the measure ofperceived usefulness.
Intention to Use
Intention to use is the willingness of knowledge workers to use a
knowledge system during work activities. The intention to use construct is
measured using five item-level questions, each with a five-point Likert scale:
response 1 for strongly agree, response 5 for strongly disagree, and response 3
as undecided or neutral. The intention to use item-level variables are measured
on the interval scale. Frequency tables and mean, median, mode, and standard
deviation statistics were used to describe the characteristics of the intention to
use variables. Item analysis (Cronbachs Alpha) and inter-item correlation tests
were conducted to measure relationships between each item-level variable and
to estimate the reliability of the measure of intention to use.
Contributing Factors
The contributing factors questions measure the relative importance of
that factor in contributing to the respondents use of the knowledge system.
Each question is measured on a four point Likert-type scale: Very Important,

Somewhat Important, Not Very Important, Not at All Important. I describe the
characteristics of each question using frequency tables.
Each contributing factor emerged during the pilot studies when I
evaluated qualitative data collected from open-ended questions. My purpose for
including these questions was to gather richer supportive data, data which could
offer insight into an individuals level of technology use relative to their
perception of the technologys usefulness and their intent to use the knowledge
system as an information source.
The survey instrument, found in Appendix A, consists of three parts.
Part I requests demographic data on the following variables; gender, age, job
category, years of computer experience, experience using SCEL, and average
weekly SCEL use. Categorical variables such as gender, age, job category, and
computer experience are common in survey research (Marcinkiewicz, 1993 and
1994). Numerous research studies: Chau (1996); Davies, Bagozzi, and
Warshaw (1989) for example, also measure the amount of technology-specific
experience possessed by participants. Part II contains questions on Level of
Technology Use, Perceived Usefulness, Intention to Use, and Factors
Contributing to Your Use of SCEL.

The Level of Technology Use question was derived by merging
properties of three models: Level of Technology Implementation (LoTi)
(Moersch, 1995), the Technology Infusion Model (Reiber & Welliver, 1989),
and the Levels of Use of the Innovation (LoU) model (Hall, Loucks, Newlove,
& Hall, 1976). Previous Level of Use of the Innovation research used structured
interviews with branching techniques to determine which of eight levels were
appropriate for that individual. The only reliability estimate available for LoU is
an inter-rater reliability of .90 estimate gathered by Hall and colleagues (1976).
No reliability estimates are available for the LoTi model although it aligns
conceptually with LoU. Moersch reported that the model was in the process of
being validated across the nation, however no results have been published as of
this study. Marcinkiewicz (1993 and 1994) conducted research studies using the
Technology Infusion Model. Marcinkiewicz reported criterion-related validity
using Cohens kappa of .72 and a reliability estimate of .96 using the Coefficient
of Reproducibility. The Level of Technology Use ranking is a new model. No
validity and reliability estimates exist for this model. This is a significant
limitation of this study.
Questions used to measure both perceived usefulness and intention to use
variables were derived from the Technology Acceptance Model (Davis,
Bagozzi, & Warshaw, 1989). Reliability and validity estimates exist for both
variables. Perceived usefulness received reliability scores of .92 (Davis,

Bagozzi, & Warshaw, 1989), .93 (Chau, 1996), and .96 (Szajna, 1996).
Intention to use received reliability scores of .80 (Chau, 1996) and .90 (Davis,
Bagozzi, & Warshaw, 1989). Szajna reported the correlation between intention
to use and perceived usefulness is significant (r = .12, p < .01).
Chau (1996) conducted three tests to estimate convergent validity: item
reliability, construct reliability, and average variance extracted. Item reliability
estimates for perceived usefulness items ranged from .71 to .82. Item reliability
estimates for intention to use items ranged from .87 to .89. Chau reported
construct reliability estimates of .85 for perceived usefulness and .82 for
intention to use. Chau reported average variance extracted estimates of .77 for
perceived usefulness and .87 for intention to use. Chau reported the
discriminant validity estimate, using Chi Square tests of association, was
significant (= 4.01 ,p < .05).
Part HI contains open-ended questions and statements designed to gather
additional descriptive data. The initial set of statements are, I would use SCEL
in the following situations: followed by, I would not use SCEL in the
following situations:. The objective of these statements is to determine what
inhibits or enables your use of SCEL. The next set of questions asks, What
additional positive factors influence your use of SCEL? and What additional
negative factors influence your use of SCEL?. These questions allow the
respondent to add to the Contributing Factors items. The last set of statements

directs the respondents to think back to when they first used SCEL, Describe
how you used it then and Describe how you use it now. Responses to these
statements should provide additional insights into how people could progress up
a Level of Technology Use model. The final question asks for any additional
Initially email aliases were populated with names systematically selected
from an alphabetized email directory. Both the questionnaire and Invitation to
Participate were attached to an email message (Appendix E) summarily
explaining the purpose of the correspondence. Upon receipt of the responses,
the survey scores were entered into SPSS for Windows database. As each
survey was received, the respondents email address was removed from the
aliases. After one week, a follow up email message was sent to non-
respondents. Two additional reminders were sent in hopes of soliciting a
response level of 40%. The text responses to my open-ended questions were
captured in Microsoft Word documents for analysis and inclusion in Chapter 4.
After analyzing data from both open-ended and close-ended questions, I
conducted telephone interviews to collect additional data for questions that
emerged during data analysis. The interviews served both confirmatory and
exploratory purposes. The interviews confirmed data collected from the surveys

and allowed me to gather additional data concerning themes that emerged from
qualitative data analysis.
Data Analysis
Inferential statistics allow researchers to reach conclusions about
populations by using data collected from samples (Spatz, 1993). The following
table describes the inferential statistics used to analyze the relationships of
importance to my research questions.

Variable 1 Variable 2 Statistical Tests

Age Level of Technology Use Chi Square Crosstabulation
Age Perceived Usefulness' Oneway ANOVA Tukey
Age Intention to Use Oneway ANOVA, Tukey
Gender Level of Technology Use Chi Square Crosstabulation
Gender Perceived Usefulness Oneway ANOVA, Tukey
Gender Intention to Use Oneway ANOVA, Tukey
Job Category Level of Technology Use Chi Square Crosstabulation
Job Category Perceived Usefulness Oneway ANOVA, Tukey
Job Category Intention to Use Oneway ANOVA Tukey
Years of Computer Experience Level of Technology Use Chi Square Crosstabulation
Years of Computer Experience Perceived Usefulness Oneway ANOVA Tukey
Years of Computer Experience Intention to Use Oneway ANOVA Tukey
Table 3-1 Inferential Statistics Analyses
Table continued on next page.

Variable 1 Variable 2 Statistical Tests
Months of SCEL experience Level of Technology Use Oneway ANOVA, Tukey
Months of SCEL experience Perceived Usefulness Pearson Correlation
Months of SCEL experience Intention to Use Pearson Correlation
Average Weekly SCEL Use Level of Technology Use Oneway ANOVA, Tukey
Average Weekly SCEL Use Perceived Usefulness Pearson Correlation
Average Weekly SCEL Use Intention to Use Pearson Correlation
Level of Technology Use Perceived Usefulness Oneway ANOVA, Tukey
Level of Technology Use Intention to Use Oneway ANOVA, Tukey
Perceived Usefulness Intention to Use Pearson Correlation
Table 3-1 Continued
Three assumptions exist for the oneway ANOVA test: the residual errors
are assumed to be random and independent, homogeneity of variance, and the
distribution of residual errors is normally distributed. My major concern is
homogeneity of variance due to the large number of unequal ns present in these
groups. If violated, homogeneity of variance is a serious problem; if the larger
ns are associated with the smaller variances, a liberal test results. While, if the
larger ns are associated with the larger variances, a conservative test results
(Lomax, 1992). The Levenes test was conducted to determine if the assumption
of homogeneity of variance was violated. When the ANOVA test reports a

statistically significant F ratio, the Tukey post hoc test was used to determine
which pair-wise comparisons reported statistically significant mean differences.
Questionnaires were distributed to 615 employees, a sample drawn from
employees of Sequent Computer Systems. Sequents email system was used to
distribute the questionnaire and to collect participant responses. Two hundred
and three (203) employees responded to the questionnaire. Data collection
involved capturing both qualitative data from the closed-ended questions and
qualitative data from the open-ended questions. Data analysis involved using
both descriptive and inferential statistics to analyze the quantitative data and
data transformation and coding to analyze the qualitative data. Inter-rater
agreement tests were conducted to establish percent of agreement measures.
Descriptive statistics were used to present the characteristics of all variables
under investigation, both qualitative and quantitative. Inferential statistics were
used to examine relationships between demographic and technology acceptance
variables, relationships that formed the core of this studys research questions.

A purpose of this study was to examine relationships between perceived
usefulness, intention to use, and level of technology use, defined conceptually as
how extensively individuals integrate a technology into their work activities.
This study also examines certain cultural and environmental factors that could
impact technology acceptance and implementation, factors that emerged from
qualitative data analyses.
In the following sections, I present descriptive and inferential statistics
that address my research questions. Frequency tables are presented for all
variables under investigation. Mean, median, mode, and standard deviation
statistics are calculated for variables measured on interval and ratio scales. To
answer research questions on relationships, three types of statistical tests were
conducted: Chi-square tests of association, oneway ANOVAs, and Pearson
Questionnaires were distributed to a sample of 615 employees
systematically drawn from a population of approximately 2,500 employees.
Forty-three questionnaires were not delivered due to bad email addresses. Of

the remaining sample of 572 employees, 203 responded yielding a response rate
of 35.5%. Although a response rate of 50% is considered desirable for survey
research (Babbie, 1991), the 35% response rate obtained in this study is typical
for studies using mailed questionnaires.
Level of Technology Use
Results in this section are intended to answer the research question;
Which Level of Technology Use describes how extensively knowledge workers
use knowledge systems?
Frequency Percent Cumulative Percent
1 Non-Use 5 2.5 2.5
2 Utilization 49 24.1 26.6
3 Infusion 86 42.4 69.0
4 Integration 27 13.3 82.3
5 Expansion 36 17.7 100.0
Total 203 100.0
Table 4-1 Frequency Table Level of Technology Use
Level of Technology Use measures how extensively individuals have
integrated SCEL into their work activities. Referring to Table 4-1, 2.5% of the
respondents reported non-use. Several of the five respondents, when asked why
they didn't use SCEL, replied with "the information is not relevant to my job"
and "I used an earlier version of SCEL but when Sequent implemented the
newer version of SCEL, the searching mechanisms were too clumsy so I quit
using it."