Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
An Analysis of the Acceptance of File Sharing Technologies
By Music Consumers
Donald L. Amoroso
Appalachian State University
Boone, NC 28607
+001.828.262.2034
[email protected]
Yi (Maggie) Guo
University of Michigan, Dearborn
Dearborn, MI 48126
+011.313.593.5578
[email protected]
Abstract
The Internet provides for an incredible simplification
and cost reduction of music distribution. However, the
music industry has placed the blame on free music
downloading on the Internet and on other computer
technologies of music reproduction for the continued
decrease in record sales. On the other hand, although
abundant research has examined the online consumer
behavior, few studies investigate the usage of Peer-to-Peer
(P2P) file sharing technology by music consumers. In order
to understand better this phenomenon of music
downloading, the authors of this paper took the perspective
of music consumers and conducted a survey of university
students concerning their habits with respect to music
downloading and its impact on buying compact disks (CDs).
The proposed research model extended the Technology
Acceptance Model (TAM).
Data analysis partially
supported the model.
1. Introduction
The Internet provides for an incredible
simplification and cost reduction of music
distribution. However, it is sometimes believed to be
the cause of some serious problems facing the music
industry. Over the last few years, the dollar value of
recorded music sales has declined in the USA and in
the world [20]. The music industry has placed the
blame on free music downloading on the Internet and
on other computer technologies of music
reproduction. On the other hand, although abundant
research has examined the online consumer behavior,
few studies investigate the usage of Peer-to-Peer
(P2P) file sharing technology by music consumers. In
order to understand better this phenomenon of music
downloading, the authors of this paper took the
perspective of music consumers and conducted a
survey of university students concerning their habits
with respect to music downloading and its impact on
buying compact disks (CDs).
The need for this research is included in previous
work by Amoroso and Koster [6], [7], [8], [9], [23]
where the statistics related to the sales of music by
age of consumer and type of music was reported by
RIAA in 2005 and showed respondent demographics
in the areas of retail buying, expected downloading
patterns, and lost income to music record companies.
Downloading is differentiated from file sharing in
this research. Downloading is defined as the physical
movement of digital files from a computer, server,
network or other media to a local file repository. File
sharing is defined in this research as the movement of
digital files from and to file repositories, where
“servers” may or may not be present. This research
focuses on the sharing of files across and among
computers, focusing on music digitized files.
The Technology Acceptance Model [7], [12], [13],
[43] has been demonstrated to be a plausible model
explaining individual adoption and usage of
information technologies. It has been applies in areas
such as individual usage of Internet services and
customers acceptance of Internet shopping. Based on
a comprehensive literature review, we suggest an
extended model that includes antecedents and
outcome measures that many studies did not include.
In the model, external variables pertaining to music
downloading using file sharing technology are
included in addition to TAM variables. Those
variables are: age, gender, education, musical buying
patterns, type of Internet connection, and previous
experience with P2P technology.
0-7695-2507-5/06/$20.00 (C) 2006 IEEE
.
1
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
Buying
Pattern (BP)
H6a
H4a
Gender
Perceived
Usefulness (PU)
H8a
H8a
Connection
H2
Behavioral
Intention (BI)
H1a
Age
H4a
Perceived Ease of
Use (PEOU)
H6b
H3
Actual Use
(AU)
H1b
H9
H5b
H5a
Previous
Experience (EX)
H7a
H7a
Education
Figure 1. Research Model of Acceptance of File Sharing Technology
A survey was conducted among university
students with total 396 respondents. All the measures
in the study were derived from previously validated
instruments. Measures for the construct involved all
have acceptable reliability and evidence of factorial
validity. Correlations among constructs were as
expected. Structural Equation Model technique was
used as the major data analysis method. The
proposed model was partially supported. Implication
to the industry and future research are discussed.
2. Research model
The Technology Acceptance Model (TAM) [14] is
a specific adaptation of theory of planned behavior
(TPB, [3]) to understand user adoption behavior of
information technology and has been widely applied
and empirical supported. In TAM, two important
beliefs that impact individual’s attitude and intention
are perceived ease of use (PEOU) and perceived
usefulness (PU). Perceived ease of use is the degree
to which an individual believes that using a particular
system would be free of physical and mental effort.
PEOU has been found to influence usefulness,
attitude, intention, and actual use.
Perceived
usefulness is based on expectancy theory which is
concerned with an individual’s beliefs in the decision
making process [41]. PU is the degree to which an
individual believes that using a particular system
would enhance his or her performance. It has been
found that the relationship between PU and usage is
strong and consistent.
Attitude toward using is the user’s evaluation of
the desirability of his or her using the system and the
individual’s positive or negative feelings about
performing the target behavior. However, there are
mixed results regarding its effect on behavior
intention [13], [30], [40]. Although presence of
attitude makes is more parallel to TPB, reasons for
dropping this variable are: 1) Empirics have indicated
that attitude does not mediate as predicted [13]; and 2)
Attitude masks the effects of the key theoretical
constructs, perception of usefulness and ease of use,
comprising the parsimony without gain in ability to
explain [42]. Behavioral intention (BI) is a measure
of the strength of one’s intention to perform a
specified behavior. Venkatesh and Davis [43] report
that behavioral intention is a good predictor of actual
usage of a technology which has received numerous
empirical supports from prior studies.
Hypotheses 1-3
In this paper, we present a research model that
examines the propensity of music consumers to adopt
P2P file sharing technologies. Based upon the
empirical research of TAM constructs, the proposed
model (Figure 1) studies the impacts of these
constructs on the consumer adoption patterns. The
modified TAM includes six external variables (age,
buying patterns, type of Internet connection, gender,
experience using P2P, and education) that have been
shown empirically to effect the consumers’ propensity
to adopt.
First, we test following hypotheses directly derived
from TAM:
2
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
Hypothesis 1: Perceived ease of use (PEOU) is
positively correlated to the perceived usefulness (PU)
and the consumer’s behavioral intention (BI) to
download music.
Hypothesis 2: Perceived usefulness is positively
correlated to the consumer’s behavioral intention to
download music.
Hypothesis 3: Behavioral intention to download
music is positively correlated to the consumer’s
actual downloading of music (AU).
Hypothesis 4
In addition to the core TAM constructs, gender and
experiences have been studies often in TAM
literature. Studies [18], [45] have reported differences
between men and women on their perception and
decision regarding information technology.
Song
and Walden [74] conclude that gender is significant
with respect to the effects of information cascades and
network externalities in consumer adoption of P2P
technologies. Specifically, female likelihood of the
adoption decision is correlated with the relative level
of adoption (network externalities). Since P2P
technologies are new and user interfaces are relatively
more complex, we hypothesize that women will rate
the perceived ease of use lower than men.
Hypothesis 4: Gender will influence the perceived
usefulness and perceived ease of use.
H4a: Women will rate the perceived usefulness of
P2P technologies lower than men will.
H4b: Women will rate the perceived ease of use of
P2P lower than men will.
Hypothesis 5
Next hypothesis concerns the influence of
experience on user’s perceptions and ultimate
behavior. Ajzen and Fishbein [4] demonstrate that
prior experience is a determinant of behavior. Many
studies have established a positive relationship
between experience with computing technology and
other constructs, such as perceived ease of use [2],
[14], and intention to use [39], [46]. In contrast, the
data show that in 2000, 2001, and 2002, downloading
music is one of the few activities for which
experience does not play a key role; it is more or less
an equally popular pastime among new users and
seasoned veteran users alike [27]. However, in 2003
report, 59% music down-loaders have experience over
three years [28]. Thus, we would like to test
following hypothesis:
Hypothesis 5: Experience (EX) will influence the
perceived ease of use and the behavioral intention to
download music using P2P technology.
Hypothesis 6
Hypothesis 6 is an extension of the TAM, with a
focus on the consumer's buying patterns. Customer
buying patterns are differentiated by their preference
for file sharing one side of the product continuum and
their preference for CD music on the other of the
continuum [36]. Consistent with the criticism of the
industry, Snir found that buying patterns will affect
the actual downloading of music. The greater the
propensity of the consumer to prefer using file sharing
technologies and Internet-based music over CD
music, the greater the perceived usefulness of file
sharing technology. Amoroso and Koster [9] find that
people, who buy music online, also buy retail, burn,
and return. There are strong correlations with
consumers who downloaded music to preview songs
and those who plan on spending more on music,
either in retail stores or online. In addition, strong
correlations between the behavior of downloading
songs to preview and downloading music to the harddrive and burning are found. While examining the
bundling and distribution of music on the Internet,
Altinkermer and Bandyopadhyay [5] proposed a
model that analyzes the demand functions of different
classifications of buyers. They also report that the
music industry turmoil is created as a result of a
general lack of understanding of the consumer’s
propensity to utilize P2P technologies, and their
aversion to high music pricing.
Hypothesis 6: Buying patterns (BP) will influence
the perceived usefulness and the actual downloading
of music.
H6a: Consumers that have a greater propensity
toward file sharing and Internet music will have a
greater perceived usefulness of P2P technologies.
H6b: Consumers that have a greater propensity
toward file sharing and Internet music will download
music to a greater extent.
Hypothesis 7
Another often studied individual difference
regarding technology acceptance is levels of
education. The effect of education has usually been
posited and found to be positive on the technology
adoption [2], [33]. However, in the context of music
downloading, the possibility of a negative correlation
between level of education and the specific
information
technology
of
Internet
music
downloading is discovered by Amoroso and Koster
[9] while performing an analysis of file sharing on the
3
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
Internet, and in a recent report by Madden and
Lenhart [27], where it is determined that 23% of
online college graduates are downloading music files,
as compared to 31% to 39% of Internet users with
lower levels of education. In fact consumers that are
more educated with respect to the ethical impacts of
downloading music illegally will most likely not use
P2P technologies. To test this possible link we
propose that education will be negatively correlated
with both the intention to download and the actual
user downloading behavior, as described in
hypothesis 7.
Hypothesis 7: Level of education will influence
both the behavioral intention to download music and
the actual downloading of music.
Hypothesis 8
Studies have reported links between age and
technology adoption [11]. However, Rogers [33]
concludes that about half of the studies on the subject
show no relationships. One example cited is that of
Sindi [35] who found no direct effect between age,
attitudes and intentions toward using an expert
system. Further empirical results designed directly
for measuring age and its influence on the TAM has
not been forthcoming. There are, however, several
published surveys that report demographic results for
Internet users involved in the downloading of music
files. The older an Internet user, the less likely he or
she is to have downloaded music [27], [31]. Similar
results can be found in a market research [22]
surveying 7,688 Internet users around the world.
Amoroso and Koster [9] report similar findings with
63% of respondents indicating they downloaded
music to their hard drive and later burned a CD for
listening. Of the respondents, 92% were between 18
and 33 years of age. Therefore, it is obvious that age
plays a role with respect to Internet users who actually
download music files. What is less clear is how age
influences the primary constructs of the TAM. With
these facts in mind, we attempt to extend the TAM
with the inclusion of age as an influencing variable as
described in hypothesis 8 and posit that the factor
affect the adoption intention by affecting the two
antecedents.
Hypothesis 8: Age will influence the perceived
usefulness and perceived ease of use for downloading
music.
Hypothesis 9
Since music files are considerable larger,
intuitively, it is easy to assume that the faster the
Internet connection (greater download speed), the
greater the perceived ease of use of Internet music
downloading technologies. However, there is little
empirical evidence to support this assumption. In
fact, most Internet data sources fail to ask questions
about the types of Internet connections [16]. One
study of factors affecting e-commerce adoption by
Lee and Park [24] reports that only 42% of their
survey respondents had Internet connection speed
exceeding 56kbps. This result was similar to that of
Rose, Khoo, and Straub [34]. However, it is found
76% of survey respondents had Internet access speeds
greater than 56kbps in [9]. The differences may be
explained in other survey demographic differences
(survey population biases), or the increased
availability of broadband Internet access between
2001 and 2003. Correspondingly, Madden and
Lenhart [24] report that 41% of Internet users with a
broadband connection at home have ever downloaded
music versus a quarter of dial-up users. Similar
results are found in [28]. Of additional interest is a
study by Xia and Sudharshan [44], who report that
interruptions
that
limit
online
consumers’
concentration reduce Web users’ satisfaction with
online shopping. Davison and Cotton [14] reported
that the strongest Internet activities associated with
type of Internet connection are downloading music,
paying bills online, and banking. Overall, those with
broadband connections are significantly more likely
to spend longer on the Internet than those with dial-up
connections. Their findings were consistent with
those of Horrigan and Rainie [17]. In an attempt to
clarify this issue we present hypothesis 9.
Hypothesis 9: The type of Internet connection will
influence the perceived ease of use of music
downloading technologies.
3. The Method
Data were collected via a survey of university
students concerning their habits with respect to music
downloading and their future music buying behavior.
The survey is presented in the appendix. The TAM
scales of perceived usefulness, perceived ease of use,
and behavioral intentions were adapted from [12] and
[15].
Perceived usefulness of peer-to-peer file
sharing included measuring the ability to accomplish
the downloading of music easier, improve the
efficiency of downloading music, increase the
likelihood of success in downloading music, and
provide alternatives to purchasing music CD's at
retail. Perceived ease of use measured the easiness of
learning to download music on the Internet, easiness
to obtain the desired music file, whether the processes
for downloading music are clear and understandable,
ease of file sharing, ease of becoming skillful at using
4
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
music downloading programs, and overall ease of
downloading technologies in general. Behavioral
intention to download music via the Internet was
measured as a combination of carrying out the
downloading task and planned use of the downloaded
music files [1], [12], [7]. Measures of behavioral
inclinations now and in the future were also used [7],
[44]. The actual usage construct was measured as a
perceived level of use of P2P downloading
technologies, knowing that there are conflicting
opinions on the reliability of such self-reported
measures [38], [39], [40], [414). The level of use
during one week time frame was used in conjunction
with relative downloading frequency and the
likelihood of adopting P2P technology for
downloading music. These measures were chosen in
an attempt to minimize the potential problem with
common-method variance, as suggested by [44].
The external variables used in the study included
experience using P2P, buying patterns, gender, age,
education and type of Internet connection. The
experience using P2P construct was examined by
measuring the perceived experience using music
downloading technologies, coupled with the number
of years using P2P to download music on the Internet
[25], [43], [44]. We examined buying patterns with
measurements of frequency of online versus offline
purchase of CD's [5], the tendency to buy retail, burn,
and return ([9], and the interest in use of fee charging
Internet music downloading services (per use, per
song, subscription). As suggested by Gefen and
Straub [18], gender was measured as a single item
(female or male). Age was measured as a single item
using a Likert type scale of 1 to 5, for five age groups.
Three were narrowly defined (2 = 19-20, 3 = 21-22, 4
= 23-24) and two were broadly defined (1 = ≤18, 5 =
≥25). The age variable range was based upon earlier
studies of Internet users who had downloaded music
files [8], [27], and given that the survey sample
population was primarily university students. We
examined education with a single item measurement
similar to that of the age variable. The primary
education level of interest was that of the
undergraduate, based upon previous research [9],
[27]. The type of Internet connection was a single
item measurement.
4. Data Analysis and Results
Total 396 university students participated in the
study, 195 male and 200 female students. Majority
participant were undergraduate (89%) within an age
range from 19 to 22 (85%). 80% of the participants
enjoyed high speed Internet connection (Cable/DSL,
T1, or better), which was higher than national average
of 60% [31]. Most of them started downloading
music when they were in high school (60%) or in
college (20%). Some records had missing values. Six
records were excluded because most data were
missing. We used mean replacement for those
records with only a few missing values. All other
constructs had acceptable reliability, shown in Table
1.
Table 1. Factor Reliability
Construct
Perceived
Usefulness (PU)
Perceived Ease of
Use (PEOU)
Behavioral
Intention (BI)
Previous
Experience of
Using (EX)
Actual Usage (AU)
Buying Patterns
(BP)
Number
of Items
5
Standardized
Cronbach’s
alpha
.840
6
.928
9
.831
5
.861
6
8
.808
.669
Our next step of data analysis was to test the
proposed model using Structure Equation Modeling,
which is a confirmatory approach and will allow us to
test the hypothesized relationships simultaneously. A
two-step approach was employed in model
construction and testing [9]. First, the measurement
model was assessed to see if any structural model
existed that had acceptable goodness-of-fit using a
confirmatory factor analysis model with covariances
between all pairs of factors. Our base model (CFA1)
included the latent factors and measured variables for
all indirectly unobservable factors: perceived
usefulness (PU), perceived ease of use (PEOU),
behavioral intentions (BI), actual usage (AU), buying
patterns (BP), and experience (EX).
Maximum likelihood estimation was used to fit the
model with the data. Although many tests of
statistical significance and indices of fit help in the
evaluation of model fit, “there is ultimately a degree
of subjectivity and professional judgment in the
selection of a ‘best’ model” [29]. In our assessment
of the models, we gave priority to the RMSEA with
considerations of other fit indices, following the
suggestions of MacCallum [26].
As Table 2 shows, overall goodness-of-fit for this
initial model was reasonable in that RMSEA was
equal to .068 (with a confidence interval of .065 to
.072) and CFI was .869. Examining the regression
5
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
weights of indicator variables on latent factors
revealed that no indicators of behavioral intention
loaded on the factor. A closer look on the items
showed that question 1 and 9 were worded differently
in term of tense from all other questions.
Deleting this two items (resulting in Model CFA2)
improved reliability even further to .91 and regression
weights became significant. Several items also caught
our eyes based on examination of the modification
indices for covariances among error terms.
We examined the covariances between AU2 and
AU4 usage and between BI2 and BP1 in the third
model (CFA3). Modification indices suggested a
cross-loading of AU5 on BI. Again, we looked at the
item and found out the item was worded in term of
future likelihood of using the technology. Thus, we
moved AU5 to be one of the indictors of BI’s (Model
CFA4).
Lastly, BP3 had no significant loading on BP;
dropping this indicator further improved the model.
CFA5 has acceptable goodness-of-fit and formed the
measurement model for our next step of analysis.
Table 2. Measurement Model Modification
D Ȥ 2/df NNFI NFI CFI RMSEA
F
CFA1 1953.50 687 2.844 .843 .793 .855
.068
Model
Ȥ2
BP
.014
PU
Gender
.068
Delete BI1 and BI9 from the base model CFA1
CFA3 1570.87 612 2.567 .878 .830 .888
.063
Free some error covariances from CFA2
CFA4 1511.12 612 2.469 .886 .836 .895
.359*
-.062
-.056
Age
-.059
Connec
tion
.595*
A
BI
.618*
-.150
.031 .017
PE
.013
.545*
The base model
CFA2 1733.54 614 2.823 .858 .812 .869
-.131**
-.053
Education
.671*
EX
*. Significant at .01 level
**. Significant at .05
l l
Figure 2. The Structural Model (base)
.061
Move AU5 to BI from CFA3
CFA5 1324.33 576 1.736 .904 .855 .912
.058
Delete BP3 from CFA4
After the “purifying” process, we fit the research
model and refined the model by dropping low tstatistics paths (based on the Wald test) and adding
parameters to the model when appropriate (based on
the Lagrange Multiplier test). Using theory as
primary guide, every change made to the model was
based on the consideration of interpretability and
improvement in fit indices. The resulting model
should have a better overall fit and R-square statistics,
as well as a theoretically reasonable structure. First,
we fitted the model proposed in Figure 1. The model
(Figure 2) has reasonable fit (Ȥ2=1557, df=711, Ȥ 2/df
= 2.189, CFI=.90, RMSEA =.055). However, several
proposed paths were not significant. After dropping
non-significant paths, we arrived at the revised model
(Figure 3). The revised model also had reasonable
fitness (Ȥ2=1415.10, df=584, Ȥ 2/df = 2.423, CFI=.90,
RMSEA =.060).
PU
BP
.298*
.636*
-.124**
.596*
AU
BI
PEOU
.467*
.696*
EX
*. Significant at .01 level
**. Significant at .05 level
Figure 3. The Structural Model (revised)
6
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
5. Discussion
Based on our structural equation model analysis,
we summarized the hypotheses in Table 3.
Significant paths represented supported hypotheses.
From the results, we can see that most of TAM
hypotheses (H 1-3) were supported as expected,
except the effect of perceived ease of use on
behavioral intention. An ad hoc model with only
paths in the core TAM model was run against the
data, revealing all paths were significant. Overall, our
study showed the validity of TAM.
Among the individual difference factors we have
examined, only experience and buying pattern had
expected effects. Age, gender, education, and Internet
connection type had no effects on user’s perception of
ease of use and usefulness of downloading music
using P2P technology.
Table 3. Hypotheses Results
Hypotheses
H1
H2
H3
H4
H5
H6
H7
H8
H9
Results
Partially supported
Supported
Supported
Not supported
Supported
Partially supported
Not Supported
Not supported
Not supported
However, previous experience has significant
effects on ease of use and behavioral intention. That
means, the more experience the user had before, the
stronger the intentions are to continue downloading
music in the future. Buying pattern was thought to
have effects on perceived usefulness and actual usage,
however, our analysis revealed only significant path
between buying pattern and actual usage.
Additionally, buying pattern showed negative path
coefficient, which means a relationship contradictory
to previous study and our hypothesis. It seems like if
the consumers are willing to pay for the music one
way or another (e.g., buys online or in retail, pay for
membership or subscription), they would use file
sharing as the means to obtain music less.
Since music industry has accused users that access
music downloading and copying made not only
possible but easy by the Internet and file sharing
technology, it is interesting to have our study results
interpreted. First, it seems if the consumers are
already experienced in downloading music via P2P
technology, they are unlikely to stop in the future.
However, our research did not ask what made them
start downloading music in the past.
On the other hand, the actually usage of file
sharing for music downloading had a negative
correlation with buying pattern. If consumers are
willing to pay for the music one form or another, they
would download music illegally less. The question
then facing the industry is how to encourage
customers to buy, either provide better products (extra
materials in albums), and/or at better price. More
dramatic, the industry should make use of the
technology at their advantage rather against it. They
can setup their own P2P file sharing network to
provide consumers better services, previewing songs,
paying by songs, membership to access of huge
volume of music, and so on. There are signs that
legitimate digital music market has taken off [21].
However, we have to be aware of the limitation of
the study, especially in understanding the
implications. The sample involved in the study was
rather narrow: college students. Lack of variance in
sample data might be the reason of insignificant effect
of age, education, and connection on perceived
usefulness and ease of use. Our research supported
the use of TAM in explaining music consumers’
behavior using file sharing technology. We shall
continue the research by including a more diverse
sample.
Since music downloading for free is an issue
involving ethical and normal judgment, and even
legal consequences, which makes it special compared
to other information technology adoption. User’s
perception of ethical aspect of this issue and possible
legal consequences it might bring are deemed to have
impact on consumer behavioral since subjective norm
is an important force in individual behavior and is part
of the theory of planned behavior [4]. Survey data
[32] state that one of seven Internet users stopped
downloading music because of recent legal actions
taken by music industry and the awareness of
copyright and ethical issues. A more recent report
also shows a decrease in Internet music infringing
using file sharing technology in term of number of
users and value [21]. The effects of these two are
worth looking into in the future.
6. Conclusion
This research effort tried to understand the
determining factors of music consumer utilizing file
sharing
technology.
It
related
individual
characteristics and user perceptions to actually
behavior of music downloading, which was thought
as a threat by music industry. We found that
perceived usefulness is related to the consumer’s
behavioral intention to download music. Also, we
7
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
found that behavioral intention was strongly related to
actual use. This confirms the original Davis model
relating attitude, behavior, and actual usage of
technology, specifically related to downloading music
on the Internet. We found that experience will
influence the ease of use and behavioral intention to
download music using P2P technology. We found
partial support for buying patterns and the influence
on perceived usefulness of P2P music downloading
technologies. Surprisingly, we did not find any
significant relationships between education, age and
actual usage, or with the type of Internet connection
and perceived ease of use.
Thus, the research contributed to our knowledge of
music downloading behavior from the perspective of
consumers. Our survey shed some light on the
chances of success of commercial music downloading
and what its features should be. P2P technology is
powerful and has become widely used by Internet
users. We believe the technology can be used to the
advantage of both consumers and music industry.
[9]
7. References
[14]
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Agarwal, R. and Karahanna, E., “Time Flies
When You Are Having Fun: Cognitive
Absorption and Beliefs About Information
Technology Usage,” MIS Quarterly, 24 (4) 2000,
pp.665-694.
Agarwal, R. and Prasad, J., “Are Individual
Differences Germane to the Acceptance of New
Information Technologies?” Decision Sciences,
30 (2), 1999, pp.361-391.
Ajzen, I., "The Theory of Planned Behavior,"
Organizational Behavior and Human Decision
Processes, 50, 1991, pp.179-211.
Ajzen, I. and M. Fishbein., Understanding
Attitudes and Predicting Social Behavior,
Englewood Cliffs: Prentice-Hall., 1980
Altinkemer, K. and Bandyopadhyay, S.,
“Bundling and Distribution of Digitized Music
over the Internet,” Journal of Organizational
Computing and Electronic Commerce, 10 (3),
2000, pp.209-224.
Amoroso, D. L. (2003). Managing the Next
Wave of Digitized Products over the Internet: An
Analysis of Music Consumers in a Rapidly
Changing Industry. Society of Information
Management (SIM) Academic Conference, 10.
Amoroso, D. and Gardner. C, “Development of
an Instrument to Measure the Acceptance of
Internet Technology by Consumers,” HICSS,
2004.
Amoroso, D. and A. Koster. “The Impact of
Internet Distribution and File Sharing on the
Music Industry,” European Applied Business
Research Conference, Venice, Italy, May, 2003.
[10]
[11]
[12]
[13]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
Amoroso, D. and Koster, A., "An Analysis of File
Sharing On The Music Industry," Working Paper,
College of Business Administration, Department
of Information & Decision Systems, San Diego
State University, 2003.
Anderson, G. C. and Gerbing, D., “Structural
Equation Modeling in Practice: A Review and
Recommended Two-Step Approach,”
Psychological Bulletin, 103 (3), 1988, pp.411423.
Brandcheau, J. C. and Wetherbee, J. C., “The
Adoption of Spreadsheet Software: Testing
Innovation Diffusion Theory in the Context of
End-User Computing,” Information Systems
Research, 1 (2), 1990, pp.115-143.
Chau, P.Y.K., "An Empirical Assessment of a
Modified Technology Acceptance Model,"
Journal of Management Information Systems
13(1), 1996, pp.185-204.
Chau, P. and Hu, P., “Information Technology
Acceptance by Individual Professionals: A Model
of Comparison Approach,” Decision Sciences, 32
(4), 2001, pp.699-719.
Davis, F.D., "Perceived Usefulness, Perceived
Ease of Use, and User Acceptance of Information
Technology," MIS Quarterly, 13(3), 1989,
pp.319-340.
Davis, F., Bagozzi, R., and Warshaw, P., “User
Acceptance of cComputer Technology: a
Comparison of Two Theoretical Models,”
Management Science, 35, 1989, pp.982-1003.
Davison, E. and Cotton, S. R., “Connection
Discrepancies: Unmasking Further Layers of the
Digital Divide,” First Monday, 8 (3), 2003,
Available [Online]:
http://www.firstmonday.dk/issues/issue8_3/davis
on/, (June 8, 2005).
Dishaw, M. and Strong. D., “Extending the
Technology Acceptance Model with Task
Technology Fit Constructs,” Information and
Management, 36 (1), 1999, pp.9-21.
Gefen, D., and Straub, D. “Gender Differences in
the Perception and Use of E-mail: an Extension to
the Technology Acceptance Model,” MIS
Quarterly, 21 (4) 1997, pp.389-400.
Horrigan, J. and Raine, L., “The Broadband
Difference,” Pew Internet and American Life
Project, June 2002, Available [Online]:
http://www.pewinternet.org/reports/toc.asp?Repo
rt=63, (December 28, 2003).
IFPI, "IFPI Releases World Recorded Music
Sales Figures for First Half of 2004," 2004,
Available [Online]: http://www.ifpi.org/sitecontent/press/20040930.html, (June 8, 2005)
IFPI, "Digital Music Report 05," 2005, Available
[Online]: http://www.ifpi.org/sitecontent/library/digital-music-report-2005.pdf,
(June 8, 2005)
8
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
[22] Ipsos-Reid, “File Sharing and CD Burners
Proliferate,” June 2002, Available [Online]:
http://www.ipsosna.com/news/pressrelease.cfm?id=1542, (June 10,
2005)
[23] Koster, A. & Amoroso, D. L. (2003). Evolution
of Music Distribution over the Internet. San
Diego International Systems Conference.
[24] Lee, D. and Park, J., “On the Explanation of
Factors Affecting e-Commerce Adoption,”
Twenty-Second International Conference on
Information Systems, 2001, pp. 109-119.
[25] Legris, P., Ingham, J., and Collerette, P., “Why
Do People Use Information Technology? A
Critical Review of the Technology Acceptance
Model,” Information & Management, 40, 2002,
pp.191-204.
[26] MacCallum, R., "Commentary on Quantitative
Methods in I/O Research," The IndustrialOrganizational Psychologist (35:4), 1998
[27] Madden, M., and Lenhart, A. "Music
Downloading, File-Sharing and Copyright," Pew
Internet and American Life Project, July 2003,
Available [Online]:
http://www.pewinternet.org/reports/pdfs/PIP_Cop
yright.Memo.pdf, (June 8, 2005)
[28] Madden, M., and Rainie, L., "America’s Online
Pursuits: The Changing Picture of Who’s Online
and What They Do," Pew Internet & American
Life Project, December 2003, Available [Online]:
http://www.pewinternet.org/pdfs/PIP_Online_Pur
suits_Final.PDF, (June 8, 2005)
[29] Marsh, H. W. and Jackson, S. A., "Flow
Experience in Sport: Construct Validation of
Multidimensional, Hierarchical State and Trait
Responses," Structural Equation Modeling, 6 (4),
1999, pp.343-371.
[30] Mathieson, K., “Predicting User Intentions:
Comparing the Technology Acceptance Model
with the Theory of Planned Behavior,
Information Systems Research, 2 (3), 1991,
pp.173-191.
[31] Rainie, L., and Horrigan, J., "Internet: The
Mainstreaming of Online Life," Pew Internet and
American Life Project, Jan 2005, Available
[Online]:
http://www.pewinternet.org/pdfs/Internet_Status_
2005.pdf, (June 8, 2005)
[32] Rainie, L., Madden, M., Hess, D., and Mudd, G.,
"The State of Music Downloading and FileSharing online," Pew Internet Project And
Comscore Media Metrix, April 2004, Available
[Online]:
http://www.pewinternet.org/pdfs/PIP_Filesharing
_April_04.pdf, (June 8, 2005)
[33] Rogers, E. M., Diffusion of Innovations, (4th ed.),
1995, New York, N. Y.: The Free Press
[34] Rose, G., H. Khoo and Straub, D.W., "Current
Technological Impediments to Business-toConsumer Electronic Commerce,"
Communications of the AIS 1 (16), 1999, pp.1-74.
[35] Sindi, A. A. “Acceptance of Information
Technology: User Acceptance Model for Expert
Systems,” Unpublished doctoral dissertation,
University of Michigan, 1992.
[36] Snir, E. M., “The Record Industry in an Era of
File Sharing: Lessons from Vertical
Differentiation,” 24th International Conference on
Information Systems (ICIS 2003), Seattle,
Washington, USA, December 2003.
[37] Song, J. and Walden, E., “Computer Behavior in
the Adoption of Peer-to-Peer Technologies: An
Empirical Examination of Information Cascades
and Network Externalities,” Ninth Americas
Conference on Information Systems, August
2003, Tampa, Florida, pp. 1801-1810.
[38] Straub, D., Limayem, M., and KarahannaEvaristo, E., "Measuring System Usage:
Implications for IS Theory Testing," Management
Science 41(8), 1995, pp.1328-1342.
[39] Taylor, S. and Todd, P., “Assessing IT Usage: the
Role of Prior Experience,” MIS Quarterly, 19 (2),
1995, pp.561-570.
[40] Taylor, S and Todd, P., “Understanding
Information Technology Usage: A Test of
Competing Models,” Information Systems
Research, 6 (4), 1995, pp.144-176.
[41] Venkatesh, V., "Determinants of Perceived Ease
of Use: Integrating Control, Intrinsic Motivation,
and Emotion into the Technology Acceptance
Model," Information Systems Research, 11(4),
2000, pp.342-365.
[42] Venkatesh, V., "Creation of Favorable User
Perceptions: Exploring the Role of Intrinsic
Motivation," MIS Quarterly, 23 (2), 1999,
pp.239-260.
[43] Venkatesh, V., and Davis, F.D., "A Model of the
Antecedents of Perceived Ease of Use:
Development and Test," Decision Science, 27 (3),
1996, pp.451-481.
[44] Venkatesh, V., and Davis, F.D., "A Theoretical
Extension of the Technology Acceptance Model:
Four Longitudinal Field Studies," Management
Science, 46 (2), 2000, pp.186-204.
[45] Venkatesh, V., and Morris, M.G., "Why Don't
Men Ever Stop to Ask for Directions? Gender,
Social Influence, and Their Role in Technology
Acceptance and Usage Behavior," MIS Quarterly,
24 (1), 2000, pp.115-139.
[46] Venkatesh, V., Morris, M., Davis, G., and Davis,
F., “User Acceptance of Information Technology:
Toward a Unified View,” MIS Quarterly, 27(3),
2003), pp.425-478.
[47] Xia, L. and Sudharshan, D., “An Examination of
the Effects of Cognitive Interruptions on
9
Proceedings of the 39th Hawaii International Conference on System Sciences - 2006
Consumer Online Decision Processes,” Second
Marketing Science Internet Conference, USC,
Los Angeles, California, 2000.
Appendix
Perceived Usefulness
PU1. Using file sharing (P2P) technologies can help me
download music.
PU2. Using file sharing (P2P) helps me download music
easier.
PU3. Using file sharing (P2P) technologies gives me an
option for downloading music.
PU4. Using file sharing (P2P) has decreased my spending
on retail CDs.
PU5. Using file sharing (P2P) has enhanced my
effectiveness in downloading music.
Perceived Ease of Use
PEOU1. Learning to download music online is easy for me.
PEOU2. I find it easy to download music from the Internet.
PEOU3. Processes for downloading music off the Internet
are clear and understandable.
PEOU4. I find Internet file sharing (P2P) to be easy to use.
PEOU5. It is easy for me to become skillful at using the
music download programs.
PEOU6. I find the music downloading technologies easy to
use.
Behavioral Intentions
BI1. I download music files for free to preview songs and
later purchase a retail CD.
BI2. I always try to download music via the Internet rather
than buy it retail.
BI3. I plan to download music for free from the Internet in
the future.
BI4. I intend to continue using the Internet to download
music for free in the future.
BI5. I expect my use of music download technologies to
increase in the future.
BI6. I plan to obtain music in the future by downloading it
to my harddrive directly and burning my own CDs.
BI7. I will use P2P more as the network of available music
increases.
BI8. I will increase my use of file sharing (P2P) despite the
threat of record labels suing downloaders.
BI9. I would buy retail CDs rather than download if there
was better packaging with retail CDs (including free DVD
and/or booklet).
(BI1 and BI9 were dropped during analysis.)
EX5. Number of years using P2P technologies to download
music on the Internet: _____
Actual Usage
AU1. How many music files do you download per week:
_________.
AU2. Number of sites you visit per week to preview music:
__________.
AU3. Indicate how frequently you use the Internet to
download music
AU4. How many different music Websites do you visit per
week: ________.
AU5. My likelihood of adopting P2P technology to
download music is very strong.
AU6. My recent number of downloads is _______.
(AU5 was moved to BI during analysis.)
Buying Pattern
BP1. How often do you buy CDs at the retail store?
BP2. How often do you buy CDs online?
BP3. How often do you buy CDs, burn a copy, and later
return for a refund.
BP4. I would be interested in per-use and per-song fee
online.
BP5. I would be interested in a long-term subscription for
online music for a discount.
BP6. I would be interested in becoming a member of a site
licensing.
BP7. I buy music retail when I cannot find it online.
BP8. I would be more likely to buy a retail CD than
download it if it costs $6-$8.
(BP3 was dropped during analysis.)
Experience
EX1. I have a good deal of experience using music
downloading technologies.
EX2. I have a good deal of experience accessing music
online using P2P technologies.
EX3. I do not need help downloading music from the
Internet using P2P technologies.
EX4. I have experience using my computer as a server for
others to download music.
10