Determinants of Instant Messaging Use
by
Namkee Park
Doctoral Student
Annenberg School for Communication
University of Southern California
3502 Watt Way
Los Angeles, CA 90089
Phone: 213-740-3939
Email: [log in to unmask]
• Submitted to the Communication Technology and Policy Division of
Association for Education in Journalism and Mass Communication.
ABSTRACT
Instant messaging is a technological innovation featuring near-real-time
communication and interactivity between users, which also exhibits network
effects in its diffusion. This study identifies a profile of instant
messaging users and empirically tests the prediction of network effects.
The study results indicate that technological innovativeness is the only
significant factor in predicting instant messaging use. It was also found
that network effects play a critical role for the users to choose a
specific service.
Jung-Sook Lee Competition
Determinants of Instant Messaging Use
Among a variety of the Internet applications, instant messaging (IM) has
been a powerful form of communication and one of the most popular
applications on the Internet. Instant messaging is "a text-based means of
near-real-time communication between users," with text messages popping up
immediately on a recipient's computer screen, and it allows the users to
maintain a list of people that they wish to interact with (Faulhaber, 2002,
p.314). The users can send messages to any of the people in the list, often
called a "buddy list" or "contact list," as long as that person is online.
Instant messaging differs from e-mail in that it is a true conversation,
"operating in synchronous rather than asynchronous mode" (Faulhaber, 2002,
p.314).
According to a recent Nielsen/NetRatings survey, more than 41 million, or
nearly 40% of the active Internet surfing population at home, used at least
one of instant messaging services during the month of May 2002. In
addition, nearly 12.6 million office workers used instant messaging during
the same time period, reaching 31% of the total active Internet population
at work[1] (PR Newswire, 2002a). Another survey says that almost 74% of
young people, roughly 13 million, have used an instant messaging program
while 44% of online adults have tried instant messaging at one time or
another (Lenhart, Rainie, & Lewis, 2001). In fact, instant messaging has
seen phenomenal growth, especially for young people. Statistics show that
college students access the Internet and utilize instant messaging programs
more than the overall US population. According to a survey from the Pew
Internet & American Life Project, college students are among the heaviest
users of instant messaging in the US. While only half of all Internet users
have sent instant messages, nearly three quarters of college Internet users
have done so. College Internet users are twice as likely as the average
Internet users to use instant messaging on any given day (Jones, et al.,
2002). Thus, a private research firm even predicts that instant messaging
will be used more often than email by 2005 (Henry, 2002). In this regard,
the purpose of this study is to understand the adoption and usage of
instant messaging by identifying a profile of the users among college
students along with implication of the characteristic of network effects in
the use of instant messaging.
Instant Messaging and Interoperability
It was in the mid-1990s when instant messaging was really exploded on the
Internet scene. ICQ, short for "I seek you," was created by four Israeli
programmers to improve the way people communicate with each other over the
Internet and within the first six months more than a million users
registered (Barken, 2002). Almost at the same time, America Online (AOL)
pioneered instant messaging with adding its chat function and the "buddy
list,"—a quick way for users to check if their friends or the people they
want to interact with were online—although the company had introduced an
earlier form of instant messaging to its customers in 1989. In 1999,
several companies such as Microsoft, Yahoo!, Otigo, and Tribal Voice
established other instant messaging services (Faulhaber, 2002), but
currently four services, AOL Instant Messenger (AIM), MSN Messenger, Yahoo!
Messenger, and ICQ, are competing with each other to attract more
subscribers.[2] As of May 2002, AOL Instant Messenger has about 22.1
million subscribers (21%), MSN Messenger about 15.7 million subscribers
(15%), Yahoo! Messenger about 12.4 million subscribers (12%), and ICQ has
4.4 million subscribers (4%) in the US, according to Nielsen/NetRatings
data[3] (PR Newswire, 2002a).
However, unlike e-mail services, none of these services are compatible with
each other and each service relies on a proprietary network with isolated
user communities. Since America Online has dominated instant messaging
service with a first-mover advantage, the company has refused to open its
network to other service providers indicating safety and security problem
and technical matters as the company's legitimate argument.[4] In
particular, for the past three years, America Online has been
intermittently blocking its rivals' attempts to send messages to its tens
of millions of users (Schofield, 2002). It means that a user of AOL Instant
Messenger cannot reach a user of MSN Messenger as well as a user of Yahoo!
Messenger cannot talk with a friend who has an account in AOL IM. Because
each instant messaging service cannot be interoperable, about 25% of
instant messaging users have accounts with more than one service (Emling,
2002).
Despite this non-interoperability, the number of instant messaging users
has increased due to its attractive function of quick sending and receiving
messages back and forth. Not only is it quick, but also the users can
control his/her own buddy list, or those authorized to send them a message,
and therefore can lock out spam (Faletra, 2002). Another major advantage of
instant messaging is that it enables users to be able to stay in touch with
their friends or relatives who live outside their communities. Although it
does not contain the visual and aural cues that people get in face-to-face
communication or phone contacts, talking to buddies online has become a
modern communication method with which users make interpersonal
communication a more valuable activity through immediate interactions
beyond geographical boundaries. These attributes of instant messaging have
helped it become popular and widespread among the Internet users. Moreover,
other companies than America Online have been trying to attract users with
different strategies focusing on utilizing their own advantages, and thus
accelerating the diffusion of instant messaging even without
interoperability. For instance, MSN has increasingly tied instant messaging
into its operating system. Its Windows Messenger service is a central
feature in Window XP and is interoperable with its Web-based product MSN
Messenger (Hu, 2002). In addition, MSN has tried to include video
conferencing as another key feature (Wilcox, 2001). Yahoo! also has
continued to support efforts towards functional interoperability with its
Web portal (Hu, 2002), and recently had a contract with AT&T Wireless
through which Yahoo! Messenger users on personal computers can exchange
instant messages with AT&T Wireless subscribers, even if the mobile user
does not have a Yahoo! ID (PR Newswire, 2002b). These efforts of MSN and
Yahoo! have given them tremendous growth in their instant messaging
services over the past two to three years, although non-interoperability
may limit their abilities to lock in users and market other services and
products (Hu, 2002).
Instant messaging is still in the stage of development in terms of widening
its user base. Thus, the prominent importance of instant messaging to both
users and the service providers requires more research. In particular, an
understanding of users and non-users of instant messaging in the stages of
adoption may speed up diffusion of the relatively new interactive
communication service. Based on these characteristics of instant messaging,
the research questions on which this study focuses are as follows:
RQ 1: What factors are most important in determining whether users choose
to be adopters of instant messaging?
RQ 2: What is the most important factor in selecting a specific service
among various kinds of available services if the users determined to use an
instant messaging service?
Diffusion Theory
This study assumes that instant messaging is an innovative interactive
technology in the era of Internet communication and uses diffusion theory
as its theoretical framework. Broadly speaking, diffusion theory addresses
the characteristics of innovations and those who adopt them (Atkin,
Jeffres, & Neuendorf, 1998). According to diffusion theory (Rogers, 1995),
adoption of technological innovations is a function of people's social
locators, media use patterns, uses of other technologies, and peoples'
communication needs. Another important concept is innovativeness, which
Rogers (1995) defines as "the degree to which an individual is relatively
earlier in adopting an innovation than other members of a social system"
(p.22). Put differently, the characteristics of earlier adopters may be
different from later adopters or non-adopters because of the degree of
their adoptive innovativeness. Thus, if we consider instant messaging as an
"innovation," diffusion theory may offer clues about who are relatively
early to adopt it and what their characteristics are.
Demographics
Several studies of the diffusion theory indicate that demographic variables
are associated with new technology adoption and use behaviors. This
argument has been supported by a variety of diffusion research about
personal computers (Dickerson & Gentry, 1983; Dutton, Rogers, & Jun, 1987;
Lin, 1998), the Internet (Atkin, Jeffres, & Neuendorf, 1998), HDTV
(Dupagne, 1999), and VCRs (Krugman, 1985; Reagan, 1987). These studies find
that the adopters of new technologies tend to be upscale, better educated,
and younger than non-adopters as Rogers (1995) suggests. Also, although
Rogers (1995) does not mention gender difference in adopting a new
technology, some works find that males are more likely to adopt new
technologies (e.g., Krendl, Brohier, & Fleetwood, 1989; Dupagne 1999).
A number of research, however, find that demographic differences between
adopters and non-adopters of new technologies have been few. For example,
Jeffres and Atkin (1996) discovered that income and education had only a
weak relationship with interest in adopting specific Internet utilities.
Kang (2002) also found that no demographic variables had a significant
impact on people's subscription to digital cable. Neuendorf, Atkin and
Jeffres (1998) found no significant demographic variable in the adoption of
audiotext audio information services and Atkin and LaRose (1994) presented
a similar result in the adoption of cable service. In terms of gender
difference, many empirical studies suggest that gender is not related to
communication technology adoption (Collins, Reagan, & Abel, 1983; Lin,
1998; Reagan, 1987). Even further, Cummings and Kraut (2002) present that
females are using the Internet more than males for communication as the
usage of the Internet has become matured. Considering the contrasting
findings, this study tests the following hypothesis regarding demographics:[5]
H1: There will be differences between users and non-users of instant
messaging in terms of age, gender, and disposable income per month.
Media Use
Diffusion theory suggests that earlier adopters will use media or
communication technologies more heavily than later adopters (Rogers, 1995).
But, in fact, this assumption has been rarely supported in the literature
(Dupagne, 1999). Rather, the media substitution hypothesis (Krugman, 1985;
Lin 1994a) holds that the introduction of a new technology encourages a
reorganizing in the way people use established technologies. For instance,
Vitalari, Venkatesh and Gronhaug (1985) found that computer adopters spent
less time with television and participated in fewer recreational
activities. Similarly, James, Wotring and Forrest (1995) found that the use
of electronic bulletin boards reduced time spent with television viewing,
book reading and telephone use.
In contrast, in terms of potential videotext news bulletin adoption,
newspaper reading level was found to have no effect on readers' intention
to adoption (Heikinnen & Reese, 1986). In the same fashion, Lin (1994b)
found a similar pattern of non-effects for videotext use on other media.
Jeffres and Atkin (1996) also discovered the use of online services was
generally not consistently related to use of other media technologies.
Considering these contrasting findings, the following hypothesis was set forth.
H2: Media use levels (i.e., television viewing, radio listening, newspaper
and magazine reading, and moviegoing) of instant messaging users will be
lower than those of non-users.
Communication Technology Clusters
According to Rogers (1995), "all technology cluster consists of one or more
distinguishable elements of technology that are perceived as being closely
interrelated" (p.15). It suggests that the adoption of one technology is
likely to stimulate the use of functionally similar technologies (Atkin &
LaRose, 1994; LaRose and Atkin, 1992). Reagan (1987), for instance, found
that adoption of communication technologies was powerfully related to
adoption of other technologies; such as videotext, personal computers,
compact disks and cable. Lin (1998) also noted that computer adoption was
related to Internet adoption intentions as well as a technology adoption
index (comprised of 14 communication technologies). Neuendorf, Atkin and
Jeffres (1998) applied Rogers' notion of technology clusters to the
adoption of audiotext information services, suggesting that use of the
audiotext was related to functionally similar technologies such as
videotext, ATMs, and 800 numbers. Therefore, a positive relationship is
expected between ownership of communication technology products and the use
of instant messaging.
H3: Instant messaging users will have more communication technology
products than will non-users.
Communication Needs
Research on communication technology adoption indicates that users' needs
are primary determining factors (Neuendorf, Atkin, & Jeffres, 1998). In
their investigation of primary motives for using the Internet, Papacharissi
and Rubin (2000) found that the most salient use of the Internet reflected
an instrumental orientation, which has been defined as "an active and
purposive orientation, often having to do with information seeking, and
characterized by utility, intention, selectivity, and involvement" (p.181).
As Lin (1994a) suggests, these motives represent a fundamental
psychological element. Also, a number of literature indicated that users'
needs or use patterns were more powerful than demographics in explaining
the adoption of the Internet (James, Wotring, & Harris, 1995), computers
(Perse & Courtwright, 1993), videotext (Reagan, 1987; Lin, 1994b), audio
information services (LaRose & Atkin, 1992; Neuendorf, Atkin, & Jeffres,
1998), cable (LaRose & Atkin, 1988; Reagan, 1991; Jacobs, 1995), and ISDN
(Jeffres & Atkin, 1996). Thus, the following hypotheses were posed:
H4: Instant messaging users will present a greater desire to accomplish
communication needs than will non-users.
H5: Use of instant messaging will be more powerfully explained by personal
communication needs than by demographic variables.
Innovativeness
According to diffusion theory (Rogers, 1995), technological innovation
adoption is associated with one's innovative traits to try to new products.
In fact, the causes of innovativeness have their psychological roots in an
individual's novelty-seeking motives (Hirshman, 1980) and these roots of
innovativeness include personality styles such as venturesomeness and
communication usage patterns (Foxall & Bhate, 1991). Based on this
understanding, Lin (1998) reported that computer adopter groups presented
the highest degree of need for innovativeness (e.g., willingness to learn
new ideas, willingness to explore new technology, and keeping up with new
technology) compared with likely-adopters or non-adopters. The following
hypothesis deals with people's innovativeness in the adoption of instant
messaging.
H6: Instant messaging users will perceive themselves to be more innovative
than will non-users.
Network Effects
For interactive communication technologies that are new, and that are thus
perceived as an innovation, their adoption sometimes depends on the
perceived number of others who already adopted the innovation (Mahler &
Rogers, 1999). When the value of a product or service to one user depends
on how many other users there are, this product or service exhibits network
effects or network externalities,[6] which sometimes referred to as
"demand-side economies of scale" (Shapiro & Varian, 1999, p.14). Instant
messaging is a perfect example of network effects in the sense that the
value of it can be enhanced as the number of users increases. Thus, as
mentioned above, if the instant messaging services are not interoperable
with each other, the leading service, AOL Instant Messenger, is likely to
attract more users because the service becomes more valuable to all users
with a network effect.[7] Thus, it seems reasonable that network effects
can account for the changes in instant messaging adoption and provide more
predictive power and guidance. With this reasoning, the present study tries
to empirically prove network effects in instant messaging.
In the early stages of the diffusion of an interactive innovation, the rate
of adoption may proceed very slowly. But eventually enough adopters are
reached when many individuals perceive that "everybody is doing it" (Mahler
& Rogers, 1999, p.721). At this point the pool of innovative technology
users has reached critical mass[8] and an individual considering adoption
of the innovation perceives that the innovation would have sufficient
utility to justify its adoption (Mahler & Rogers, 1999). Thus, achieving
critical mass in an innovation is a turning point from which the innovation
can be widespread with network effects.
Innovations characterized by network effects often have a winner-take-all
result, with all potential users leaning toward a single product or service
provider (Cummings & Kraut, 2002; Markus, 1987) and the users of other
products or services are likely to transfer to the leading product or
service due to the large user base. Thus, theoretically, if a dominant
provider, America Online in this study, chooses not to interconnect, other
competitors may be driven from the market. Then new entrants or smaller
competitors can face a significant barrier to entry and the leading
provider may enjoy monopoly (or near-monopoly) status.[9] A monopoly based
on network effects may be especially difficult to overcome, even with a
superior product or service (Faulhaber, 2002).
As Shapiro and Varian (1999) and Mahler and Rogers (1999) properly pointed
out, however, users' expectations or perceived number of other adopters are
more crucial to reach critical mass or to become the leading provider in an
interactive innovation. The service that is expected to become the standard
will eventually become the standard (Shapiro & Varian, 1999). Also, the
critical mass is socially constructed by individuals, based on their
communication with relevant others (Mahler & Rogers, 1999).
Nevertheless, it is hard to say that users' choice of a specific service is
determined by a single factor. Although this study assumes that network
effects are critical in choosing an instant messaging service, there can be
other factors that influence users' choice. According to consumer research
literature, consumer choices concerning the selection or consumption of
products and services can often be difficult and important to the consumer,
to marketers, and even to policy makers (Bettman, Luce, & Payne, 1998).
Thus, consumers usually compare several different brands before deciding
which option to purchase (e.g. Dhar, Nowlis, & Sherman, 1999). Consumers at
times compare products or services in order to make judgments of similarity
that are basic to categorization, generalization, and discrimination
(Nosofsky, 1986; Tversky, 1977). It is likely that the users of
communication technologies judge similarity or dissimilarity when they come
across a new innovation, instant messaging. In this study, three factors,
user satisfaction with functions and features of instant messaging
services, connectedness to others Internet services, and company image,
were employed as the alternative factors to network effects.
Given the similarity of functions of instant messaging services, feature or
layout of instant messaging services will more attract users, especially
fancy-driven young users. Even though all instant messaging services have
similar features as well as functions, appealing to consumers' cognitive
mechanisms with unique features has been a focal point for marketers to
construct users' preference (Dhar, Nowlis, & Sherman, 1999). Another
factor, connectedness to other Internet or communication services, is
currently a major marketing strategy for MSN and Yahoo!, as mentioned
above. Convenient connection to other services such as email, Web browsing,
portal and directory service, and wireless connection of the service
provider with an instant messaging service, usually called "one-stop
shopping," can attract more instant messaging users in establishing a large
user base. Finally, company image is related to brand loyalty. A company
that is perceived as the "brand-as-partner" to users and provides the same
identity with users makes the users consistently be loyal to the company's
services and feel user-brand bonds (Fournier, 1998). Considering that all
instant messaging service providers are innovative technology-oriented,
company image may be a corollary of users' innovativeness in the sense that
company image can be connected to users' innovativeness in selection of a
service. Keeping in mind these alternative factors, the following
hypothesis was posed.
H7: Instant messaging adopters' selection of a specific service will be
more explained by the perceived number of existing users than by user
satisfaction, connectedness to other services, and company image.
Method
Data Collection
This study conducted a survey using self-administered questionnaire in a
large private research university in the West Coast area from February 24
to March 7, 2003. A convenient sample of respondents was 168 Communication
and Political Science undergraduate students who were attending
introductory classes in each department. Although it is not a
representative sample for the overall population, it might be appropriate
for this study considering college students' heavier use of instant
messaging than others.
Questionnaire Design
The questionnaire focused on three dependent variables, whether or not the
respondents use instant messaging, adoption rate, and which service they
(primarily) have been using, and five independent variables for
investigating use and adoption of instant messaging, (a) demographics, (b)
media use, (c) technology clusters, (d) communication needs, (e) innovative
attitude, as well as four independent variables for examining network
effects, (f) the perceived number of other users, (g) user satisfaction
toward functions and features, (h) connectedness to other services, and (i)
company image.
Instant messaging use: There are two categories of instant messaging
adopters, users and non-users. Users were asked to report their use of
instant messaging. The dependent measure of use of instant messaging used
the following phrasing: "In the last week, how many times did you use
instant messaging?" and "In the last week, how many hours did you use
instant messaging?" The metric value of each respondent's answer was retained.
Adoption rate: The concept of adoption rate was constructed based on
Rogers' (1995) categorization. Rogers specified people's technology
adoption types as follows: "innovators" (2.5%), "early adopters" (13.5%),
"early majority" (34%), "late majority" (34%), and "laggards" (16%) along a
time dimension. Using this timeline, an adoption rate was developed and
this dependent variable was operationally measured by obtaining respondent
answers to the question, "How long have you been a user of instant
messaging service?" The responses were coded with 8 categories, from 0
(non-user) to 7 (over 5 years).[10] Thus, this dependent variable assesses
the speed of user adoption with respect to instant messaging.
Specific instant messaging service: Users were asked what kind of instant
messaging service they have been using (i.e., AIM, MSN, Yahoo!, ICQ or
other). If they use more than one service, they were asked to designate a
primary service.
Demographics: To obtain demographic data, respondents were asked about
their age, gender, and disposable income per month. Ratio scale was used
for age (years); gender was dummy coded; ordinal scales were used to
measure disposable income per month, ranging from 1(less than $100) to 6
(over $901). Considering that respondents are college students and
attending a private university, it was hard to measure their income
directly. Thus, disposable income per month was used as a substitute.
Media use: Respondents were asked to report the number of hours they spent
watching television, listening to radio, and reading a newspaper daily, in
addition to the number of times they read magazines on a weekly basis.
Frequency of moviegoing in a month was also asked.
Technology clusters: Respondents were asked whether they own or subscribe
to any of a list of ten communication technology products. They include
personal computer, broadband access, PDA, cell phone, video game player,
DVD player, digital camera, video camera, cable television subscription,
and DBS subscription. These ten items were coded as dummy variables (0=no,
1=yes). The number of items owned was then summed to reflect the extent of
each respondent's technology ownership.
Communication needs: With regard to operationalizations of communication
needs, respondents were asked how much they agreed or disagreed with three
statements. They were told to respond on a scale of 1 to 7, where 1 means
strong disagreement, 7 means strong agreement, and 4 is neutral. The need
for engaging in interpersonal communication was measured with the following
statements:[11] "I spend a lot of time talking with friends and associates
about things I find interesting, like hobbies, personal interests, or
current issues," "I often feel the need to express myself to others," and
"If there was some way I could send a message to others, I would do it
regularly" (Cronbach's a = .71).
Innovative attitude: Respondents' attitude about their own innovativeness
was measured by asking the following questions: "On a scale of 1 to 7 where
1 means not technically progressive at all, or low tech, and 7 means very
technically progressive, or high tech, how would you rate yourself?"[12] "I
enjoy trying out new technologies and like to introduce them to my friends
or colleagues," and "I consider myself a modern person who is usually
up-to-date on new technologies."[13] Again, they were told to respond on a
scale of 1 to 7, where 1 means strong disagreement, 7 means strong
agreement, and 4 is neutral, in the second and the third statements
(Cronbach's a = .88).
Network effects: With regard to operationalizations of network effects,
users were asked how much they agreed or disagreed with the following
statement: "I think the instant messaging service which I'm currently using
has the largest number of users among instant messaging services." They
were told to respond on a scale of 1 to 7, where 1 means strong
disagreement, 7 means strong agreement, and 4 is neutral.
User satisfaction: User satisfaction toward functions and feature of
instant messaging services was measured by asking four questions of how
much users like and satisfy with the functions and features of the instant
messaging service they are (primarily) using. A 4-point Likert scale
ranging from 1 (not at all) to 4 (very much) was used for the
questions (Cronbach's a = .92).
Connectedness to other services: Connectedness to other services was
measured by asking how regularly users of an instant messaging service are
using the email, Web browsing, directory service of the instant messaging
provider. A 4-point Likert scale ranging from 1 (never) to 4 (always) was
used for each service (Cronbach's a = .74).
Company image: Users' perception about the image of the company they are
using an instant messaging was measured by asking the following questions:
"On a scale of 1 to 7 where 1 means very bad image and 7 means very good
image, how would you rate the image of the company you are using the
(primary) instant messaging service?" and "On a scale of 1 to 7 where 1
means not technically innovative at all and 7 means very technically
innovative, how would you rate the company you are using the (primary)
instant messaging service?" (Cronbach's alpha a = .72).
Data Analysis
To test the study's hypotheses with respect to instant messaging use and
choice of a specific service, logistic hierarchical regression analyses
were performed because the dependent variables are categorical. Also,
hierarchical multiple regression was used to assess the relative influence
of the independent variables in predicting respondents' adoption rate. In
order to screen potential mulitcollinearity problems with the predictor
variables for the predictive equation involving adoption rate, Pearson's
correlation coefficients were computed for all independent variables. The
highest intercorrelation among the independent variable was .37, suggesting
that serious multicollinearity problems probably did not exist with respect
to the estimated regression model.
Results
Descriptive Results
The sample was 42.3% male and 57.7% female students. The mean age of the
sample was 20.15 (SD = 1.90) and the median disposable income per month was
in the category of $200 to $400. Of all respondents, 79.8% has used at
least one of instant messaging services and among the non-users, 23.5%
planned to use a service in the near future. The median hours of instant
messaging use per week was 4 and half hours and 41.4% of the users have
used instant messaging over 5 years.
Among the users 78.4% has used AOL IM for their instant messaging service
or as their primary service in case they use multiple services. In
addition, 43.3% of the users have been using more than one instant
messaging service.
Logistic Hierarchical Regression – Users of Instant Messaging
The result by logistic hierarchical regression for the users of instant
messaging is presented in Table 1. The logistic hierarchical regression
examined the relative influence of demographics, media use, ownership of
technology products, communication needs, and technological innovativeness
in predicting instant messaging use. Demographics (age, gender, and
disposable income) were entered first, followed by media use (television
viewing, radio listening, newspaper and magazine reading, and moviegoing),
ownership of technology clusters, communication needs, and technological
innovativeness. The Wald statistic, the equivalent of the t test in linear
regression, was used to determine the statistical significance of the
regression coefficients (Knoke, Bohrnstedt, & Mee, 2002). The improvement
chi-square (?2) test, which is comparable to an F-change test in linear
regression, tested the null hypothesis that coefficients for the variables
added at each step of the regression were 0 (Dupagne, 1999).
Table 1. Logistic Hierarchical Regression: Instant Messaging Users vs.
Non-Users
Predictors
Step Entered
Improvement
?2 Test
-2 Log Likelihood
?
Age
Gender
Disposable income
Media use
Technology cluster
Communication needs
Innovativeness
1
2
3
4
5
6
7
2.035
2.385
.003
.021
7.531**
4.239*
14.630***
167.20
164.82
164.82
164.79
157.26
153.02
138.39
-.118
.459
-.175
-.003
.173
.248
.774***
*p<.05. **p<.01. ***p<.001.
According to Table 1, no demographic variables were found to have a
significant impact on whether respondents chose to use instant messaging.
In other words, the users of instant messaging were not significantly
different from non-users in terms of age, gender, and disposable income.
Thus, H1 was not supported.
Contrary to expectations, none of media use, technology cluster, and
communication needs was found to have a significant influence on the
instant messaging use. Interestingly, however, media use was negatively
related to whether respondents use instant messaging, although the
relationship was not significant. Therefore, H2 to H5 were not supported.
The only significant variable in this analysis was innovativeness towards
new technologies (p<.001), which supported H6, indicating that those who
use instant messaging are more likely to have innovative attitude accepting
and using new technologies. According to this model, the percentage of
classification accuracy was 82.7%.
Hierarchical Regression – Adoptive Innovativeness Results
Table 2 provides the relative influence of each variable in predicting
adoptive innovativeness (i.e., the speed of user adoption) with respect to
instant messaging. The regression model estimate yielded two significant
predictors regarding adoptive innovativeness: innovativeness ( = .302) and
technology ownership ( = .175). The findings suggest that users who have
innovative attitude towards new technologies and who have more technology
products will adopt instant messaging service earlier than other users.
Surprisingly, no other variables made a significant contribution to the
variance explained. A total of 18.2% of the variance was explained after
all of the predictor variables were entered into the regression equation,
F(7, 160) = 5.072, p<.001.
Table 2. Hierarchical Multiple Regression: Predictors of Adoptive
Innovativeness
Predictors
Step Entered
R
R2
R2 Change
ß
Age
Gender
Disposable income
Media use
Technology ownership
Communication needs
Innovativeness
1
2
3
4
5
6
7
.098
.134
.158
.161
.311
.331
.426
.010
.018
.025
.026
.097
.109
.182
.010
.008
.007
.001
.071**
.012**
.073***
-.072
.021
-.022
-.030
.175*
.047
.302***
*p<.05. **p<.01. ***p<.001.
Logistic Hierarchical Regression – Network Effects Results
The logistic hierarchical regression uncovered that the perceived number of
other users, connectedness to other Internet services, and innovative
attitude, which was included and significant in the first logistic
regression, were all significant predictors of choosing AOL IM as the
users' instant messaging service (or their primary service) as shown in
Table 3. Interestingly, however, the predictor of connectedness to other
service is negatively associated with the users' choice of AOL IM. It means
that the users of AOL IM do not heavily use other AOL services although
they use AOL IM. Among the significant predictors, the perceived number of
other users was the strongest in predicting the users' choice of a specific
instant messaging service, which support H7. In addition, the percentage of
classification accuracy was 85.1%.
Table 3. Logistic Hierarchical Regression: AOL IM Users vs. Non-AOL IM Users
Predictors
Step Entered
Improvement
?2 Test
-2 Log Likelihood
ß
Innovative attitude
Perceived number of other users
User satisfaction
Connectedness to other services
Company image
1
2
3
4
5
7.511**
34.386***
.519
13.622***
.037
132.48
98.09
97.57
83.95
83.91
.785**
1.068***
.554
-.943**
-.127
*p<.05. **p<.01. ***p<.001.
Discussion
This study profiled instant messaging users by comparing instant messaging
users to non-users in terms of demographics, media use, technology
ownership, communication needs, and innovative attitude towards new
technologies. The findings, however, only partly support hypotheses derived
from diffusion theory and from earlier empirical studies, which have
attempted to empirically test diffusion theory's application to new
technologies innovation. No significant impact of demographics on instant
messaging use can be explained by the fact that college students may have
similar characteristics in terms of social locators. Surprisingly enough,
however, communication needs were found not to have a significant impact on
the usage of instant messaging in this study. In particular, because
communication needs are considered to be the primary determining factor in
communication technology adoption (Neuendorf, Atkin, & Jeffres, 1998) as
well as fundamental psychological elements in information seeking (Lin,
1994a), the finding that communication needs were not related to instant
messaging use is unusual. Perhaps because currently almost three quarters
of young people have already been active users of instant messaging
(Lenhart, Rainie, & Lewis, 2001) as described earlier, respondents'
communication needs are less relevant as a decision factor with respect to
instant messaging use. It is also possible that innovative attitude may
dominate other variables in predicting instant messaging use especially for
college students considering they are likely to have similar
characteristics in terms of innovative technology adoption. As Lin (1998)
noted, high degree of innovativeness was the compelling factor in comparing
adopters and non-adopters for a new technology adoption.
As for the communication technologies ownership, this study partly
confirmed that past findings that early adoption of communication
technologies was related to adoption of other similar technologies (Atkin &
LaRose, 1994; Reagan, 1987; Lin, 1998; Neuendorf, Atkin, & Jeffres, 1998).
It is perceivable, however, that the impact of technology ownership may be
diminished due to relative easier access to the Internet in campus although
it was a significant factor in the early stage of instant messaging
diffusion. It can explain that technology ownership was a significant
factor in predicting adoption rate while it was not in instant messaging
use. As a theoretical finding, the results of this study propose that once
an innovative communication technology has been widely used or achieved
critical mass, communication needs may not play a critical role in
predicting its use. Rather, innovative attitude can be the decision factor
in adopting the technology, as reported here.
The results associated with network effects were more interesting.
Consistent with network effects' prediction, results of this study indicate
a high likelihood of AOL's dominance in instant messaging market and
empirically support the theoretical explication. Given the importance of
the critical mass in explaining the adoption of interactive innovations,
AOL IM may become of increased utility to the users, who use instant
messaging to communicate with more and more others. However, not most
important but noticeable result is that the association between the users'
choice of AOL IM and connectedness to other Internet services was
negatively related, meaning that AOL's lead in instant messaging does not
guarantee the company's control in other services, which can give room for
other competitors. It also proves that the business models of MSN or Yahoo!
to overcome AOL's refusal to interoperability are empirically meaningful.
Thus, efforts of MSN and Yahoo! should be more focused on developing
related services with instant messaging in the future, in which the next
generation of instant messaging will appeal the users with a rich array of
content from a wide variety of sources beyond the current text-based one
such as music and video streaming or video conferencing in the broadband
environment.
The results of this study also have implication for policymakers. This
study suggests that interoperability is essential for the users of instant
messaging to fully appreciate the innovative communication technology
whatever logic the leader in the marketplace employs. As reported earlier,
43.3% of instant messaging users have been using more than one service due
to the non-interoperability between the services. Policymakers need to
specify future plan in the deployment of the next generation instant
messaging,[14] which will enhance the users' welfare with respect to easier
and richer communication access as well as promoting fair competition among
the service providers.
Limitations and Future Research
As already noted, the sample used in this study was from a single
university. The percentage of instant messaging use was somewhat high
compared to a national sample and market share of each service provider was
quite different from the result of this study, which can reduce external
validity of this study. Also, the relative low scale reliability among the
variables of communication needs may weaken their predicting power.
Furthermore, this study utilized a limited number of predictive variables,
which accounted for only 18.2% of the total variance explaining instant
messaging adoption. Future research should use a more representative sample
and include more potential predictors such as computer competence or
individual communication pattern to provide a fuller explanation of the
factors accounting for instant messaging use. In addition, regarding the
influence of network effects, future study may employ a more elaborated
model to understand the relationship between the perceived number and
actual number of other users and its impact on the users' choice of a
specific service provider.
References
Atkin, D., Jeffres, L.W., & Neuendorf, K.A. (1998). Understanding Internet
adoption as
telecommunications behavior. Journal of Broadcasting &
Electronic Media, 42(4),
475-490.
Atkin, D. & LaRose, R. (1994). An analysis of the information services
adoption
literature. In J. Hanson (ed.). Advances in Telematics (Vol. 2,
pp.91-110). New
York: Ablex.
Barken, L. (2002 June). Instant messaging risks and rewards. e-Business
Advisor.
Beckwith, R. (2002). Plugged in: Instant messaging goes corporate. Newsday,
August 6.
Bettman, J.M., Luce, M.F., & Payne, J.W. (1998). Constructive consumer choice
processes. Journal of Consumer Research, 25, 187-217.
Collins, J., Reagan, J., & Abel, J.D. (1983). Predicting cable
subscribership: Local
factors. Journal of Broadcasting, 27(2), 177-183.
Cummings, J.N. & Kraut, R. (2002). Domesticating computers and the
Internet. The
Information Society, 18(3), 221-231.
Dhar, R., Nowlis, S.M., & Sherman, S.J. (1999). Comparison effects on
preference
construction. Journal of Consumer Research, 26, 293-306.
Dickerson, M.D. & Gentry, J.W. (1983). Characteristics of adopters and
non-adopters of
home computers. Journal of Consumer Research, 10, 225-235.
Dupagne, M. (1999). Exploring the characteristics of potential
high-definition television
adopters. Journal of Media Economics, 12(1), 35-50.
Dutton, W.H., Rogers, E.M., & Jun, S.H. (1987). Diffusion and social
impacts of
personal computers. Communication Research, 14, 219-250.
Emling, S. (2002). Upstarts send AOL a message: Companies vie for slice of
instant-chat
action. The Atlanta Journal-Constitution, June 9.
Faletra, R. (2002). AOL is stalling the advancement of instant messaging among
businesses. Computer Reseller News, September 16.
Faulhaber, G. (2002). Network effects and merger analysis: Instant
messaging and the
AOL-Time Warner case. Telecommunications Policy, 26(5,6), 311-333.
Fournier, S. (1998). Consumers and their brands: Developing relationship
theory in
consumer research. Journal of Consumer Research, 24, 343-373.
Foxall, J.H. & Bhate, S. (1991). Psychology of computer use: XIX. Extent of
computer
use: Relationship with adaptive-innovative cognitive style and
personal
involvement in computing. Perceptual and Motor Skills, 72, 195-202.
Heikinnen, K. & Reese, S.D. (1986). Newspaper readers and a new information
medium.
Communication Research, 13, 19-36.
Henry, S. (2002). The instant-mess age: 'IM' isn't private, and that's a
problem for firms,
workers. The Washington Post, July 21.
Hirschman, E.C. (1980). Innovativeness, novelty seeking, and consumer
creativity.
Journal of Consumer Research, 7(3), 283-295.
Hu, J. (2002). AOL not sold on IM interoperability. News.Com, July 23,
Available:
http://news.com.com/2100-1023-945909.html?tag=rn.
Jacobs, R. (1995). Exploring the determinants of cable television
subscriber satisfaction.
Journal of Broadcasting & Electronic Media, 39(2), 262-274.
James, M.L., Wotring, C.E., Forrest, E.J. (1995). An exploratory study of
the perceived
benefits of electronic bulletin board use and their impact on
other communication
activities. Journal of Broadcasting & Electronic Media, 39(1),
30-50.
Jeffres, L.W. & Atkin, D. (1996). Predicting use of technologies for
communication and
consumer needs. Journal of Broadcasting & Electronic Media,
40(3), 318-330.
Kang, M.H. (2002). Digital cable: Exploring factors associated with early
adoption.
Journal of Media Economics, 15(3), 193-207.
Katz, M. & Shapiro, C. (1994). Systems competition and network effects.
Journal of
Economic Perspectives, 8(2), 93-115.
Krendl, K.A., Brohier, M.C., & Fleetwood, C. (1989). Children and
computers: Do sex-
related differences persist? Journal of Communication, 39(3), 85-93.
Krugman, D. (1985). Evaluating the audiences of the new media. Journal of
Advertising,
14(4), 14-19.
LaRose, R. & Atkin, D. (1988). Satisfaction, demographic and media environment
predictors of cable subscription. Journal of Broadcasting &
Electronic Media,
32(4), 403-413.
LaRose, R. & Atkin, D. (1992). Audiotext and the re-invention of the
telephone as a mass
medium. Journalism Quarterly, 69, 413-421.
Lenhart, A., Rainie, L., & Lewis, O. (2001). Teenage life online: The rise
of the instant-
messaging generation and the Internet's impact on friendships
and family
relationships. Pew Internet & American Life Project, Available:
http://www.pewinternet.org/reports/pdfs/PIP_Teens_Report.pdf.
Lin, C.A. (1994a). Audience fragmentation in a competitive video
marketplace. Journal
of Advertising Research, 34(6), 1-17.
Lin, C.A. (1994b). Exploring potential factors for home videotext adoption.
In J. Hanson
(ed.). Advances in Telematics (Vol. 2, pp.111-121). New York: Ablex.
Lin, C.A. (1998). Exploring personal computer adoption dynamics. Journal of
Broadcasting & Electronic Media, 42(1), 95-112.
Mahler, A. & Rogers, E.M. (1999). The diffusion of interactive communication
innovations and the critical mass: The adoption of
telecommunications services by
German banks. Telecommunications Policy, 23(10,11), 719-740.
Markus, M.L. (1987). Toward a "critical mass" theory of interactive media:
Universal access,
interdependence and diffusion. Communication Research, 14(5),
491-511.
Neuendorf, K.A., Atkin, D., & Jeffres, L.W. (1998). Understanding adopters
of audio
information innovations. Journal of Broadcasting & Electronic
Media, 42(1), 80-
93.
Nosofsky, R.M. (1986). Attention, similarity, and the
identification-categorization
relationship. Journal of Experimental Psychology: General, 115,
39-57.
Papacharissi, Z. & Rubin, A.M. (2000). Predictors of Internet use. Journal of
Broadcasting & Electronic Media, 44(2), 175-196.
Perse, E.M., Courtright, J.A. (1993). Normative images of communication
media: Mass
and interpersonal channels in the new media environment. Human
Communication
Research, 19, 485-503.
Pindyck, R.S. & Rubinfeld, D.L. (2000). Microeconomics. (5th ed.). Upper
Saddle River,
NJ: Prentice Hall.
PR Newswire, (2002a). Instant messaging used by more than 41 million home
Internet
surfers. PR Newswire, June 17.
PR Newswire, (2002b). Yahoo! Messenger goes mobile with AT&T Wireless. PR
Newswire, October 15.
Reagan, J. (1987). Classifying adopters and nonadopters for technologies
using political
activity, media use and demographic variables. Telematics and
Informatics, 4, 3-
16.
Reagan, J. (1991). Technology adoption: Is satisfaction the best predictor?
Journalism
Quarterly, 68, 325-332.
Rogers, E.M. (1995). Diffusion of innovations. (4th ed.). New York: Free Press.
Schofield, J. (2002). Web giants battle over instant messaging access. The
Irish Times,
August 23.
Shapiro, C. & Varian, H. (1999). Information rules. Cambridge, MA: Harvard
University
Press.
Tversky, A. (1977). Features of similarity. Psychological Review, 84, 327-352.
Vise, D. (2002). AOL testing a link to ICQ: Message services may be
combined. The
Washington Post, October 30.
Wilcox, J. (2001). Microsoft targets AOL with windows messenger. CNET News,
June 4.
Available: http://news.cnet.com/news/0-1003-200-6180034.html.
[1] Media Metrix reports that nearly 72.9 million users in the US used
instant messaging services during the same time period (Beckwith, 2002).
Although there exists discrepancy between data, it is true that the number
of people using instant messaging is rapidly increasing.
[2] Although America Online acquired ICQ in 1998, the company has operated
it as a separate business (Vise, 2002).
[3] Media Metrix, however, reports that AOL IM has 31.4 million
subscribers, MSN Messenger 29.1 million subscribers, Yahoo! Messenger 19.1
million subscribers, and ICQ has 8.1 million subscribers, as of April 2002
(Emling, 2002).
[4] This interoperability issue was a condition of the merger between
America Online and Time Warner in 2001. The Federal Communication
Commission (FCC) approved the merger mandating interoperability for future
generations of the AOL's instant messaging service. But, the present
text-based service may not be applied to this condition. For more details,
see Faulhaber (2002).
[5] Considering the sample of this study, which is from college students,
I assume that there is no difference in education. Also, disposable income
per month was used as a substitute for income. For more details, see
"Method" section.
[6] For more detailed economic description of network effects, see Pindyck
and Rubinfeld (2000) Ch. 4, and for a review of the extensive literature on
network effects, see Katz and Shapiro (1994).
[7] In contrast, mobile phone does not exhibit a network effect although
it is an interactive innovation, because mobile phone adopters can connect
to the existing base of all telephone users (Mahler & Rogers, 1999).
[8] Critical mass can be defined as "the minimal number of adopters of an
interactive innovation for the future rate of adoption to be
self-sustaining" (Mahler & Rogers, 1999, p.721).
[9] This is the very reason why America Online has refused to open its
network of instant messaging to competitors and the Federal Trade
Commission (FTC) and the Federal Communication Commission (FCC) were
concerned about the merger between America Online and Time Warner.
[10] This timeframe was constructed considering that America Online, the
leader in instant messaging service, introduced the "buddy list" in 1996,
thereafter instant messaging became popular rapidly, and the company
allowed for non-AOL subscribers to free-download AIM as a stand-alone
package in 1997.
[11] These statements were used in Jeffres and Atkin (1996) and Atkin,
Jeffres and Neuendorf (1998). This study employed revised forms of the
statements.
[12] This statement was used in Kang (2002).
[13] These statements were used in Jeffres and Atkin (1996) and Atkin,
Jeffres and Neuendorf (1998).
[14] The conditions regarding the next generation instant messaging were
not clearly determined at the time of the merger between AOL and Time
Warner. For more details, see Faulhaber (2002).
|
