See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/263581208

A three-tier ICT access model for intention to participate online: a

comparison of developed and developing countries

Article  in  Information Development · April 2014

DOI: 10.1177/0266666914529294

CITATIONS READS

36 1,313

3 authors, including:

Luke Younghoon Chang Myeong Cheol Park

Beijing Institute of Technology Korea Advanced Institute of Science and Technology
63 PUBLICATIONS   1,843 CITATIONS    74 PUBLICATIONS   2,828 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

privacy in online world View project

Information Privacy and Personal Data Protection View project

All content following this page was uploaded by Luke Younghoon Chang on 02 July 2014.

The user has requested enhancement of the downloaded file.

Article
Information Development
1–17
A three-tier ICT access model for ª The Author(s) 2014
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
intention to participate online: DOI: 10.1177/0266666914529294
idv.sagepub.com
a comparison of developed and
developing countries

Younghoon Chang
Korea Advanced Institute of Science and Technology (KAIST)

Siew Fan Wong

Sunway University

Myeong-Cheol Park
Korea Advanced Institute of Science and Technology (KAIST)

Abstract
The digital divide between countries is a critical phenomenon separating the countries that ‘have’ (mostly
developed countries) from the countries that ‘have not’ (mostly developing countries). It places develop-
ing countries in a disadvantaged position economically, politically, and socially that in turns affects the life
and social well-being of the citizens. Current understanding of the digital divide is limited by usage of sec-
ondary data, which prompts scholars to call for more theoretically-based empirical research. Since the
key to reducing the digital divide is to increase people’s access to ICT, this study developed and empiri-
cally tested a 3-tier ICT access model to determine the effect on ICT development of intention to par-
ticipate in online activities among citizens in developed and developing countries. The model was founded
on three existing research frameworks. In addition, this study theoretically clarified the relationship
between four types of ICT access and how they contribute to intention to participate online. Survey data
of university students from Korea and Cambodia was collected. The results confirm the effect of ICT
development on citizens’ intention and isolate key factors that impact the two countries differently. Tech-
nological access was found to impact extrinsic motivational access and skills access. Social access influ-
enced extrinsic and intrinsic motivational access and skills access. Implications for research and
practice are discussed.

Keywords
digital divide, ICT access, online participation, ICT development, cross-national comparison, Cambodia, Korea

A country’s development index plays an important role in affecting the digital divide.

Introduction et al., 2012). Since 2007, the International Telecommu-

The digital divide is a critical, complex, and dynamic nication Union (ITU) has consistently monitored ICT
global phenomenon (Van Dijk and Hacker, 2003, Hsieh development worldwide and provides an ICT
et al., 2008). It refers to the gap between those who have
and do not have access to computers and the Internet for
Corresponding author:
a wide variety of activities (Van Dijk, 2006). The digital
Wong Siew Fan, Sunway University - Department of Information
divide between developed and developing countries Systems, No 5, Jalan Universiti, Bandar Sunway Petaling Jaya
comes in the form of information and communication Selangor 46150, Malaysia. T: 603 7491-8622 F: 603 5635-8633.
technology (ICT) development, access, and use (Chang Email: [email protected], [email protected]
2 Information Development

Development Index (IDI) that ranks ICT infrastructure

and uptake in 155 countries. Developed countries are
often ranked high in this index compared to developing
countries.
The digital divide places developing countries in a
disadvantaged position economically, politically, and
socially (Pick and Azari, 2008; Zhu and Chen, 2013;
Kaba and Said, 2013). It threatens the economic
development, government efficiency, social struc-
ture, and ultimately the ability of the countries to
compete in the global market (Pick and Azari, Figure 1. ICT access model (Source: Chang et al., 2012).
2008; Zhu and Chen, 2013; Kaba and Said, 2013).
Consequently, the life and social well-being of citi- Previous research (Van Dijk, 2006; Chang et al.,
zens in these countries are affected (Castells, 2009; 2012; Chen and Wellman, 2003) has identified four
Huyer, 2005; Broos and Roe, 2006; Ayanso et al., types of access (technology, motivational, skills, and
2010; Chinn and Fairlie, 2007). Being on the better social), but stops short of clearly proposing their
side of the digital divide, developed countries con- inter-relationships. Understanding access, especially
tinue to increase their lead over developing countries perceived level of access, is critical to address the
(Parker, 2001; Van Dijk and Hacker, 2003). Current issue of the digital divide. This is because successful
rapid technology advancement will only exacerbate access to digital technology is based not only on
this inequality and widen the gaps even further (Zhu actual, but also on perceived, levels of access (Van
and Chen, 2013). Dijk, 2006). The purpose of this study is to examine
The key to reducing the digital divide is having and compare the effects of ICT development on inten-
‘access’ to ICT (Van Dijk, 2006; Chang et al., 2012; tion to participate in online activities as between citi-
Chen and Wellman, 2003). In today’s networked soci- zens of developed and developing countries. It also
ety, relative differences in getting information and com- investigates the intricacies of perceived digital access
munication become decisive factors for individuals’ and how it leads to intention to participate in online
position in society on a smaller scale, and for a country’s activities. Founded on three existing research frame-
position in the world on a larger scale (Van Dijk and works (i.e., Change et al., 2012; Liaw, 2007; Wixom
Hacker 2003). Recognizing this fact, governments, and Todd, 1995), a 3-tier ICT access model was
regardless of their IDI ranking, try to formulate better developed and tested. The findings improve existing
policies to increase ICT access in an effort to narrow the understanding of the digital divide in general and
gap. Scholars acknowledge the complexity and multidi- digital access in particular, and help to inform effec-
mensionality of the digital divide issue. They recognize tive policy making and intervention.
that great amounts of effort in terms of technological,
cognitive, and social resources are required to address
the issue effectively (Van Dijk and Hacker, 2003; Hsieh Theory, research model, and hypotheses
et al., 2008). To achieve this, they need to gain better and
deeper understanding of the problem. Digital access
Current understanding of the digital divide is limited Having access to and usage of ICT is an important
to secondary data and qualitative studies (Van Dijk and part of life and the well-being of citizens in contempo-
Hacker, 2003). This leads researchers to call for more rary networked society (Selwyn, 2003). However,
empirical studies that focus on theoretical development many people lack access to ICT and are ‘excluded’
and modeling (DiMaggio and Hargittai, 2001; Jackson from technology advancement. This creates the phe-
et al. 2001; Van Dijk, 2006). With the exception of nomenon of the digital divide, which separates those
Hsieh et al. (2008, 2011), who examined digital inequal- who have access from those who have not. Following
ity between socio-economically different groups in the Van Dijk and Hacker (2003) and Chang et al. (2012),
USA, to the best knowledge of the authors, there is a we classified ICT access into four facets: technologi-
lack of theoretically-based empirical study that exam- cal access, social access, motivational access, and
ines ICT access in developing countries and compares skills access (Figure 1). These four facets are further
this issue between developed and developing countries. grouped into two general types of access based on
Chang et al: A three-tier ICT access model for intention to participate online 3

one’s locus of control (Chang et al., 2012; Hoffman Motivation can be classified into extrinsic and intrin-
et al., 2003). If control is located with the individual sic motivation. Extrinsic motivation relates to utilitar-
(i.e., the concept of ‘me’), then it is an internal access; ian outcomes. Utilitarian outcome refers to goal-
otherwise, it is regarded as an external access. Internal driven reasons and the expected utility gained from
access therefore depends totally on oneself, specifically, using ICT (Davis et al., 1992; Lin 2007; Venkatesh
on the individual’s own belief and behavior. Motiva- and Brown 2001). Intrinsic motivation is linked to
tional access and skills access are two types of inter- hedonic outcomes, which are enjoyment, pleasure,
nal access. External access must be provided by and inherent satisfaction derived from using ICT
external forces such as governments, societies, and (Davis et al., 1992; Lin 2007; Venkatesh and Brown
organizations (Chen and Wellman, 2003; Chang 2001). In information systems (IS) studies, perceived
et al., 2012). External access consists of social access usefulness has been used to measure extrinsic motiva-
and technological access. tion and perceived enjoyment is used to measure
Technological access is the most crucial factor that intrinsic motivation (Davis et al., 1992; Hsieh et al.
makes people use ICT (Chen and Wellman, 2003; Van 2011). Wu and Lu (2013) verified that dual-purposed
Dijk, 2006). It refers to the availability and accessibility IS which have utilitarian and hedonic characteristics
of tangible technology such as personal computers, soft- should use both perceived usefulness and enjoyment
ware, and Internet connections (Chang et al., 2012; to measure users’ motivation to use technology. Since
Chen and Wellman, 2003; Van Dijk, 2006). Hence, it using the Internet serves both utilitarian and hedonic
is also called physical access. For example, to be able purposes, measurement of motivation to participate in
to use the Internet, basic infrastructure such as hard- online activities should include both extrinsic and
ware, software, and network connections must be pres- intrinsic motivational access (Wu and Lu, 2013).
ent. Countries and their associated income levels affect Skills access refers to ‘‘one’s ability to use digital
technological access (Dewan and Riggins, 2005; technology’’ (van Deursen et al., 2006). Prior studies
Cuervo and Menendez, 2006; Ono and Zavodny, underscore people’s ability and confidence to use ICT
2007; Chen and Wellman, 2003; Chinn and Fairilie, as an important determinant of the digital divide (Van
2007). In developed countries, around 70.2 percent of Dijk, 2006, van Deursen et al., 2006, Chang et al.,
households have access to the Internet; in developing 2012). Perception of one’s skill level or ability to per-
countries, this rate is only 24.4 percent (ITU, 2013). In form better in certain tasks is called self-efficacy (Ban-
some developing countries, people do not even have dura, 1997, Compeau and Higgins, 1995). In the context
access to earlier forms of technology such as televisions of ICT usage, Internet and computer self-efficacy refers
and telephones (ITU, 2013). to an individual’s ability and confidence to use ICT.
Social access refers to the overall social conditions Digital skills access measures the overall informacy
(i.e., awareness, affordability, usability, and language index, which includes a test of instrumental and infor-
support of social contents) that influence one’s acces- mational skills (Van Dijk and Hacker, 2003). Instru-
sibility to digital contents (Chen and Wellman, 2003; mental skill is one’s ability to operate hardware and
Van Dijk, 2006; Chang et al., 2012). It is the extent to software while informational skill is one’s ability to
which users perceive digital contents as being avail- search for information using digital hardware and
able and affordable. Previous research (Lin and Lu, software.
2000) found that social backgrounds to content avail-
ability such as cost, number, language, and usability
affect one’s intention to use the Internet. Theoretical framework
Motivational access refers to one’s desire to use To understand the interrelationships among the four
ICT (Chang et al., 2012; Davis et al., 1992). While facets of digital access and how these access types
technological access and social access are basic lead to intention, we adopted Wixom and Todd’s
requirements to online participation, motivational (2005) integrated model of Delone and McLean’s
access is an essential underlying push. To be a part (1992) IS success model and Davis’s (1989) technol-
of the digital society, one needs to have the desire ogy acceptance model (TAM) (Figure 2), and Liaw’s
to participate in online activities. Research has shown (2007) three-tier use model (3-TUM) (Figure 3) as the
that non-users come not only from the ‘have-nots’ but theoretical foundation for this study.
also from the ‘want-nots’ (i.e., people who refuse to According to Wixom and Todd (2005), there are
participate in online activities) (Van Dijk, 2006). two types of beliefs that lead to intention to use:
4 Information Development

Figure 2. An integrated model of user satisfaction and technology acceptance (Source: Wixom and Todd, 2005).

Figure 3. Three-tier use model (Source: Liaw, 2007).

objective-based beliefs and behavioral-based beliefs. perceived system quality; the second tier is about
Objective-based beliefs are beliefs and attitudes affective and cognitive beliefs; and the third tier is
toward certain objects such as a computer or a net- behavioral intention. The first tier corresponds to
work. In ICT use, objective-based beliefs refer to Wixom and Todd’s (2005) objective-based beliefs,
information quality/satisfaction and system quality/ the second tier corresponds to behavioral beliefs, and
satisfaction. One’s objective-based beliefs will influ- the third tier is behavioral intention. Liaw (2003) fur-
ence his/her behavioral beliefs. Behavioral beliefs ther distinguished behavioral beliefs into two types:
include ease of use and usefulness. Objective-based affective (perceived enjoyment), and cognitive (per-
belief alone is a weak predictor of behavioral inten- ceived usefulness and self-efficacy).
tion because after the threshold of having physical With Wixom and Todd (2005) and Liaw (2007) ser-
resources has passed, users need other factors to ving as the theoretical foundations, we mapped Chang
drive them to use ICT. While technology-centered et al.’s (2012) ICT access model (see Figure 1) to build
researchers (e.g., Kvasny 2002; Van Dijk and Hacker a 3-tier ICT access model. External access which con-
2003) believe that material access (technological and sists of social access and technological access, which
social) is key to addressing the digital divide, prior are objective-based beliefs, while internal access,
research (e.g., Agarwal and Prasad 1997; Bhattacher- which includes motivational access and skills access,
jee 2001; Karahanna et al. 1999; Szajna 1996) shows are behavioral beliefs. Figure 4 depicts our proposed
that providing this access is only the first step. This is three-tier ICT access model and hypotheses.
when behavioral beliefs become important. Research
has found that when behavioral belief is added as a
mediator between objective-based belief and intention, External access (Objective-based beliefs)
it explains ICT use and use intention better (Fishbein Technological access. Technological access affects ICT
and Ajzen, 2005; Davis et al. 1989, Delone and usage (Chen and Wellman, 2003; Lenhart 2002; Van
McLean, 1992; Liaw, 2007; Wixom and Todd, 2005). Dijk, 2006). Previous research (Delone and Mclean,
Liaw’s (2007) three-tier use model (3-TUM) corre- 1992; Wixom and Todd, 2005) found a positive rela-
sponds well with Wixom and Todd (2005)’s inte- tionship between the quality of infrastructure (i.e.,
grated model. Founded on the Theory of Planned systems quality) and the quality of the information
Behavior (TPB), Social Cognitive Theory (SCT), and used (i.e., information quality). Information or con-
TAM, 3-TUM is used to determine user behavioral tent that is produced in digital format cannot stand
intention to use computers and the Internet. The first alone. It requires the support of proper ICT infrastruc-
tier in the model refers to users’ experience and ture before the contents can be shared and used. The
Chang et al: A three-tier ICT access model for intention to participate online 5

Figure 4. Research model for three-tier ICT access (3-TICTAM).

more available the ICT infrastructure is, the more 2013). Connections to the Internet through a 56K
people will perceive the contents to be accessible. dial-up modem and high-speed broadband, for exam-
Therefore, following the literature, we propose that ple, give totally different experiences which affect
technological access has a positive relationship with people’s desire to use ICT (Chen and Wellman,
social access in both Korea and Cambodia. 2003, Van Dijk, 2006). In developed countries, ICT
infrastructure is in much better shape than in develop-
H1: Technological access will positively influence ing countries. For example, ITU (2013) ranked Korea
social access in both Cambodia and Korea. first and Cambodia 120th in the ICT development
index. In Korea, ICT infrastructure is easily accessi-
The relationship between objective-based beliefs ble and readily available for people to use. In Cambo-
and behavioral beliefs has been established in the liter- dia, however, ICT infrastructure is scarce and may not
ature. Wixom and Todd (2005) verified the relation- be available when people want to use it. In a study of
ship between systems quality and perceived the use of a computer resource center, Taylor and
usefulness, while Liaw et al. (2006) confirmed the Todd (1995) found that not having enough computers
effect of system quality on perceived enjoyment and to meet the demands of all the students who want to
self-efficacy. Therefore, we hypothesize a positive use the computers in the center creates a barrier for
relationship between technological access and motiva- ICT use. Therefore, when technology is not readily
tional access and technological access and skills access. available and when competition for access is high,
The condition of ICT infrastructure will affect the people are less motivated to use ICT. Following this
efficiency and the volume of Internet use (ITU, argument, we hypothesize that differential level of
6 Information Development

ICT development in developed and developing coun- When Internet contents are affordable and accessible,
tries will moderate the relationship between technolo- people will have higher motivation to use them
gical access and motivational access in such a way (Chang et al. 2012). In developed countries where
that the impact will be stronger in developed countries online contents are easily accessible, people will have
than in developing countries. As will be discussed in higher motivation to use ICT. The preceding argu-
the methodology section, we chose Korea as the rep- ments suggest,
resentative of a developed country and Cambodia as
the representative of a developing country. All rele- H5: ICT development status will moderate the posi-
vant hypotheses discussed below will use the names tive relationship between social access and extrinsic
of these two countries. motivational access such that the relationship is
stronger for Korea than for Cambodia.
H2: ICT development status will moderate the posi- H6: ICT development status will moderate the posi-
tive relationship between technological access and tive relationship between social access and intrinsic
extrinsic motivational access such that the relation- motivational access such that the relationship is
ship is stronger for Korea than for Cambodia. stronger for Korea than for Cambodia.
H3: ICT development status will moderate the posi-
tive relationship between technological access and Liaw et al. (2006) found that Internet search quality
intrinsic motivational access such that the relation- affects one’s self-efficacy toward Internet search
ship is stronger for Korea than for Cambodia. engines. This means that information search and the
quality of results obtained positively influence peo-
Again, in developed countries where technology is ple’s perception of their self-efficacy. In developing
easily accessible, the people have more opportunities countries where social access is not readily available
to learn how to use ICT and become more proficient and less affordable, it is difficult for people to practice
compared to people from developing countries. This and gain enough confidence toward their online skills
increases their confidence level and improves their access. Specifically, Chang et al. (2012) found that
perception toward their own ICT skill sets. In con- people from developed countries have higher percep-
trast, people from developing countries have much tion of contents accessibility than those from develop-
lesser chance to use ICT, which reduces their famil- ing countries. We thus suggest,
iarity with the technology. Consequently, they will
have lesser confidence toward their level of skill H7: ICT development status will moderate the posi-
access. Thus, tive relationship between social access and skills
access such that the relationship is stronger for Korea
H4: ICT development status will moderate the posi- than for Cambodia.
tive relationship between technological access and
skills access such that the relationship is stronger for
Internal access (Behavioral beliefs)
Korea than for Cambodia.
Motivational access. Motivational access affects inten-
tion to use ICT (Davis et al., 1992; Wu and Lu,
Social access. As with technological access, social 2013). Some people see the utilitarian values of ICT
access is an objective-based belief that will influence and use ICT to improve productivity and increase
motivational access and skills access. Lin and Lu their work performance. For example, Microsoft
(2000) found that cost, number, language, and usabil- Excel can be used to perform additional functions
ity of website affect people’s behavior to use web- better than manual calculators. Others use ICT for
sites. Van Dijk (2006) stressed that Internet resource pure hedonic reasons to achieve pleasure and self-
accessibility determines people’s social inclusion. If fulfilling values. For example, many people play
a country has perfect technological access but no online games and chat with friends on social media
social access, then this country does not have digital because of the fun gained from performing these
inclusion (Chen and Wellman, 2003). Wixom and activities. In studying 303 individual studies of 120
Todd (2005) found that information quality has a pos- different ICT, Wu and Lu (2013) found a positive
itive effect on perceived usefulness, and Junglas et al. relationship between intrinsic and extrinsic motiva-
(2013) verified a positive relationship between socia- tion and usage and intention to use the systems. While
bility (social support) and perceived enjoyment. many use ICT, some in the category of ‘want-nots’
Chang et al: A three-tier ICT access model for intention to participate online 7

fail to see the appeal of participating in online activi- impact on their intention to participate online. Thus,
ties. For them, the utility is probably lower than the skills access should have a stronger impact for Cam-
costs (Katz and Rice, 2002). Therefore, the presence bodia than for Korea.
of motivational access is important to influence one’s
intention to participate in online activities. Accord- H10: ICT development status will moderate the pos-
ingly, we propose, itive relationship between skills access and intention
such that the relationship is stronger for Cambodia
H8: Extrinsic motivational access will positively than for Korea.
influence intention to participate in online activities
for both Cambodia and Korea. Research methodology
H9: Intrinsic motivational access will positively Measurement development
influence intention to participate in online activities
for both Cambodia and Korea.
In this study, measurement items that had demonstrated
good psychometric properties from previous studies
were adapted and modified to fit the current research
Skills access. Evidence suggests that Internet and com-
context. Items for technological access were adopted
puter self-efficacy exert positive influence on inten-
from Chang et al. (2012) and measure individuals’ per-
tion to use ICT (Compeau and Higgins, 1995; Hsu
ception toward the condition and accessibility of ICT
and Chiu, 2004; Liaw, 2007; Liaw, 2008). Self-
infrastructure. Social access is operationalized as the
efficacy is particularly critical in understanding the
extent to which online contents are affordable and avail-
digital divide (Eastin and LaRose 2000) as it deters
able to users. Again, the measurement items came from
disadvantaged groups from accessing and using ICT
Chang et al. (2012). Items for extrinsic and intrinsic
(Bishop et al, 2001, Crump and Mcllroy 2003; Hsieh
motivational access were adapted from Davis et al.
et al., 2008). In the Theory of Planned Behavior, self-
(1992) and Venkatesh and Brown (2001). Extrinsic
efficacy is a key component of perceived behavioral
motivational access measures perceptions of produc-
control, which influences behavioral intention (Ajzen,
tivity or performance in using the Internet. It has a uti-
2002). Similarly, in Self-Evaluation Theory, self-
litarian focus. Intrinsic motivational access measures
efficacy is a crucial factor that affects behavioral
perceptions of pleasure and self-fulfilling values from
intention (Kernis and Waschull, 1995).
using the Internet and has a hedonic focus. Skills
Computer and Internet self-efficacy level of
access items came from Bandura (1997) and measure
Koreans is much higher than that of Cambodians
the level of self-efficacy. We operationalized online
(Chang et al., 2012). This is supported by the World
access as an individual’s intention to access and use
IDI Skills index, which shows Korea ranks 11th while
the Internet and adapted the items from Lin (2006).
Cambodia ranks a distant 123rd. In general, Koreans
All items were measured using 7-point Likert scales.
have good access to ICT (ITU, 2013). Many younger
Table 1 shows the measurement items and the con-
generations of Koreans are exposed to ICT from a
struct definition.
very young age (Trinder et al., 2008; Chang et al.,
2012). As a result, they have acquired adequate skills
that allow them to better cope with the complexity Sample and data collection
embedded in ICT. In another words, the majority of To compare 3-tier online access between developing
them, if not all, have high ICT self-efficacy, which and developed countries, we chose Cambodia (a
minimizes the impact of skills access on their inten- developing country) and Korea (a developed country).
tion to participate in online activities. As shown by the country development index provided
In contrast, Cambodians are less skillful and have in Table 2, Cambodia and Korea are clearly posi-
less confidence in operating ICT (Chang et al., tioned at different development stages. Korea is one
2012). When Cambodians think they do not have the of 25 members of the Organization for Economic
skill sets and are not in control, they may develop Co-operation and Development (OECD), with a GNI
symptoms of computer anxiety and as a result will not per capita of US$ 22,670, and grouped in the high
be willing to use ICT (Cooper and Weaver, 2003). income category (World Bank, 2013). Its ICT devel-
Since acquiring ICT skills is seen as a privilege, those opment index is ranked top in the world (ITU,
who have gained ICT knowledge will have a higher 2013) and its network readiness index is ranked 11th
self-efficacy level, which in turn exerts stronger in the world (UNDP, 2013; WEF, 2013).
8 Information Development

Table 1. Construct definition and measurement items.

Construct Definition Items Measurement Items

Extrinsic Perception of productivity or performance EMA1 Using the Internet improves my study/work
Motivational of using the Internet performance.
Access EMA2 I find the Internet useful for my study/work.
[EMA] EMA3 Using the Internet enhances my effectiveness in
my study/work.
EMA4 Using the Internet provides me with
information that would lead to better
decisions.
Intrinsic Perception of pleasure and self-fulfilling IMA1 I find using the Internet to be enjoyable.
Motivational values of using the Internet. IMA2 I find using the Internet to be pleasant.
Access IMA3 I have fun using the Internet.
[IMA] IMA4 I find using the Internet to be exciting.
Social Access [SA] Perception toward the extent of SA1 I can afford to use online Internet content.
affordability and availability of online SA2 There is much online Internet content that I
contents am interested in.
SA3 There is much available [Korean/Cambodian]
content on the Internet.
SA4 There is much online Internet content available
for me.
Skills Access [SKA] Perception of one’s ability and confidence SKA1 I feel confident in understanding terms related
to use computers and the Internet to computers.
SKA2 I feel confident in understanding terms related
to Internet programs.
SKA3 I feel confident in troubleshooting basic
Internet problems.
SKA4 I feel confident in using the Internet to gather
information.
Technological Perception toward the availability and TA1 ICT infrastructure: The ICT infrastructure in
access accessibility of various ICT infrastructure my country is good.
[TA] TA2 Hardware: I can easily buy computer hardware
in my country.
TA3 Software: I can easily buy or download
software in my country.
TA4 Bandwidth: I am satisfied with the overall
Internet speed of my country.
Online Individual intention to participate in online INT1 I am willing to participate in online activities.
Participation activities INT2 I am willing to write on online sites.
Intention [INT] INT3 I am willing to share some information online.

Cambodia, on the other hand, is grouped in the low research context, that compares an online participation
income nation category with a GNI per capita of US$ model between a developed and a developing country.
880 (World Bank, 2013). Due to the long civil war that Prior to the full survey, a pilot survey was distrib-
finally ended in 1999, the country has many social and uted to students in local universities in both Cambodia
economic problems, such as lack of human resources, and Korea and tested to ensure that all wordings for
infrastructure, and political instability (CIA, 2013). Its measurement items were proper and had reliability and
ICT development is ranked 120th, far behind Korea. validity. The survey was translated to Khmer (official
Similarly, its rankings in the Human Development language of Cambodia) and Korean respectively by
Index and network readiness index also lag far behind professional translators. A total of 300 students from
Korea. Due to the major differences between the two a local university in Phnom Penh participated in the
countries, we contend that both fit well into our full-scale survey. After removing 12 missing or
Chang et al: A three-tier ICT access model for intention to participate online 9

Table 2. Comparisons of country development index validity (Chin, 1998; Hair et al., 2009). Table 6 shows
between Cambodia and South Korea (in World Ranking). that all item loadings of the constructs exceeded the
recommended level of 0.7, suggesting that more var-
South
Index Cambodia Korea
iance was shared between an item and its construct
than there was error variance. The composite reliabil-
ICT Development Index 2012 120th 1st ities were all greater than the recommended threshold
(ITU, 2013) of 0.7 (Hair et al, 2009) and the average variance
IDI Access sub index 2012 109th 11th extracted (AVE) exceeded the recommended score
(ITU, 2013) of 0.5 (Fornell and Larker, 1981) (Table 4). Further-
IDI Skills sub index 2012 (ITU, 2013) 123rd 1st more, the square roots of all AVEs were above 0.8,
IDI use sub index 2012 (ITU, 2013) 131st 2nd
which were much larger than all the cross-
Human Development Index 2012 138th 12th
correlations (Table 5) (Gefen et al. 2003), and each
(UNDP, 2013)
Networked Readiness Index 2013 106th 11th item loaded most strongly on its corresponding con-
(WEF, 2013) struct (Table 6). These assessment results indicate that
the measurement items have good convergent and dis-
criminant validity.
incomplete responses, we had 288 useful data points.
In Korea, 312 students from a local university in Dae-
jeon answered the full-scale survey. Again, 5 missing Structural model
or incomplete responses were removed, resulting in We tested the structural model for Korea and Cambodia
307 useful data for further analysis. independently and examined its path coefficients and R2
Table 3 presents the demographic characteristics values. Estimates of path coefficient indicate the
of the respondents. Among the Cambodian respon- strength of the relationships between the constructs
dents, 132 (45.8 percent) were males and 156 (54.2 while R2 values show the amount of variance explained.
percent) were females. Among the Korean respon- As presented in Figure 5, the models for both countries
dents, 151 (49.2 percent) were males and 156 (50.8 explained a rather significant amount of variance in the
percent) were females. All respondents used the dependent and mediating variables. Specifically, the
Internet. In Cambodia, the majority spent between model explains 25.1 percent of the variance in intention
one and two hours per week on the Internet. In for Cambodia and 21.3 percent of the variance in inten-
Korea, the majority spent more than five hours per tion for Korea. All relationships between the constructs
week on the Internet. About 25 percent of the respon- performed as expected and were significant, with the
dents from Cambodia did not own a computer. In exception of the relationship between technological
Korea, this number was only 5 percent. When it access and intrinsic motivational access for both Korea
comes to having Internet connection at home, in and Cambodia and the relationship between technologi-
Korea, almost everybody had this privilege, but less cal access and extrinsic motivational access for Cambo-
than half of the Cambodian respondents had Internet dia. Extrinsic and intrinsic motivational access and
connection at home. Despite this, a very high per- skills access were found to positively influence inten-
centage (75.7 percent) of Cambodians used social tion to participate online. Social access significantly
media. influenced extrinsic and intrinsic motivational access
and skills access. Technological access significantly
Results influenced skills access.
The measurement and structural models were estimated
using Smart-PLS version 2.0 (http://www.smartpls.de). Multi-group analysis for differences between devel-
The measurement and structural models were first eval- oped and developing countries (Korea versus
uated separately for Korea and Cambodia before cross- Cambodia)
group comparisons were conducted.
To test differential behavioral impact between Cam-
bodia and Korea, we conducted Smith-Satterthwait
Measurement model (S-S) test by analyzing the path coefficients and their
The analysis of the measurement model includes tests variance for each country (Hsieh et al., 2008). S-S test
of reliability, convergent validity, and discriminant uses pair-wise t-tests to assess and compare the
10 Information Development

Table 3. Demographic characteristics of the respondents.

Cambodia South Korea

(N ¼ 288) (N ¼ 307)

Demographic Category Frequency Percentage Frequency Percentage

Gender Male 132 45.8% 151 49.2%

Female 156 54.2% 156 50.8%
Age Under 18 8 2.8% 2 0.7%
19*24 266 92.4% 192 62.5%
25*29 12 4.2% 86 28.0%
30*34 2 0.7% 27 8.8%
Education Level Undergraduate 285 99.0% 234 76.2%
Master 3 1.0% 63 20.5%
Doctoral 0 0% 10 3.3%
What is the average time you use the Internet per Less than 30 min. 14 4.9% 0 0%
week? 30 min. * 1 hour 47 16.3% 2 0.7%
1 *2 hours 91 31.6% 11 3.6%
2 * 3 hours 69 24.0% 40 13.0%
3 *5 hours 37 12.8% 40 13.0%
More than 5 30 10.4% 214 69.7%
hours
Where do you use the Internet? Home 105 36.5% 146 47.6%
School 85 29.5% 4 1.3%
Internet Café 68 23.6% 0 0%
Others 30 10.4% 157 51.1%
Do you have a computer? Yes 214 74.3% 292 95.1%
No 74 25.7% 15 4.9%
Do you have Internet connection at home? Yes 130 45.1% 302 98.4%
No 158 54.9% 5 1.6%
Are you using social media? Yes 218 75.7% 269 87.6%
No 70 24.3% 38 12.4%

structural differences and variance between path coef- Discussion

ficients (Vinzi et al., 2009; Chin 1998). This test is The results support seven of the ten hypotheses
suitable for assessing discrete moderators and has (Table 8) and reveal four paths with differential
been used to analyze differences between countries impact between Korea and Cambodia (Table 8). This
and other groups (Hsieh et al., 2008; Vinzi et al., implies that a country’s development index plays
2009; Susanto et al., 2013). an important role in affecting the digital divide. Korea
Table 7 shows the results of path comparisons is the one of the most advanced countries in terms of
between Cambodia and Korea. Four paths (H2, H5, ICT development and diffusion, while Cambodia is
H6, and H10) were found to be different between Korea one of the countries in the lowest development group.
and Cambodia. The path coefficient from technological In line with prior studies (Davis et al., 1992; Wu
access to extrinsic motivational access for Korea was and Lu, 2013), motivational access, both intrinsic and
significantly stronger than that of Cambodia. The path extrinsic, positively influenced intention (H8 and H9
coefficients from social access to extrinsic motivational were supported). The more motivated individuals are
access and intrinsic motivation for Korea were also sig- toward using ICT, the higher the likelihood of their
nificantly stronger than the path coefficients for Cambo- joining digital societies. Similarly, skills access posi-
dia. For the relationship between skills access and tively influenced intention, with Cambodia having
intention, Cambodia showed a stronger effect than stronger influence than Korea (H10 was supported).
Korea. Overall, the analysis results provide evidence Also, as predicted, technological access had a strong
of the moderating effect of ICT development index on effect on social access for both countries (H1 was
the behavioral models of intention to participate online.
Chang et al: A three-tier ICT access model for intention to participate online 11

Table 4. Construct reliability and convergent validity.

Cambodia Korea

Construct Mean (S.D.) Composite reliability AVE Mean (S.D.) Composite reliability AVE

Extrinsic Motivation 5.75 (0.97) 0.89 0.66 5.83 (0.87) 0.94 0.80
Intrinsic motivation 5.01 (1.17) 0.90 0.69 5.48 (0.94) 0.96 0.85
Intention 4.69 (1.37) 0.91 0.76 4.72 (1.24) 0.97 0.91
Skill 4.70 (1.12) 0.89 0.67 5.22 (1.00) 0.94 0.81
Social 4.73 (1.10) 0.86 0.62 5.58 (0.89) 0.94 0.79
Technological 4.22 (1.29) 0.88 0.65 5.79 (0.81) 0.90 0.69
Note: S.D.: Standard Deviation, AVE: Average Variance Extract.

Table 5. Correlation matrix and average variance extracted.

Cambodia (N¼288)

Construct EMA IMA INT SKA SA TA

Extrinsic Motivation 0.81

Intrinsic motivation 0.51 0.83
Intention 0.37 0.34 0.87
Skill 0.39 0.35 0.44 0.82
Social 0.49 0.46 0.44 0.58 0.79
Technological 0.31 0.27 0.45 0.50 0.53 0.81

Korea (N¼307)

Construct EMA IMA INT SKA SA TA

Extrinsic Motivation 0.89

Intrinsic motivation 0.57 0.92
Intention 0.37 0.36 0.95
Skill 0.51 0.41 0.39 0.90
Social 0.63 0.62 0.43 0.65 0.89
Technological 0.48 0.39 0.29 0.54 0.52 0.83
*Diagonals represent the square root of the AVE while off diagonals represent the correlations.

supported). This shows that with the availability of motivation. For Korea, the finding was in line with pre-
necessary ICT infrastructure, the society will be able vious research that found quality and satisfaction of sys-
to generate more useful and relevant contents (Chen tems affected perceived usefulness (Wu and Lu, 2013).
and Wellman, 2003). Contrary to our expectation, technological access
The impact of technological access on extrinsic moti- did not have significant impact on intrinsic motiva-
vational access was rejected in Cambodia but supported tional access for either Cambodia or Korea (H3 was not
in Korea, with the latter having stronger effect (H2 was supported). One possible reason for this may be that the
supported). Cambodia has limited infrastructure and respondents linked ICT infrastructure in general to
device availability (ITU, 2013). Even though Cambo- potential utilitarian values, and not hedonic values –
dians can buy most ICT systems and subscribe to ICT for example, how would ICT enable them to complete
services, the prices are relatively higher than in Korea. work faster? This is especially the case in Cambodia,
Higher ICT prices (ITU, 2013) cause Cambodians to where ICT infrastructure is expensive. Citizens would
have lower purchasing power of ICT. High pricing and pay for computers and subscribe to the Internet only
the lack of ICT infrastructure nullify the impact of tech- when the devices and services could return
nological access on Cambodians’ perceived extrinsic performance-related values.
12 Information Development

Table 6. Loadings and Cross-loadings.

Cambodia Korea

Item INT IMA EMA SA SKA TA INT IMA EMA SA SKA TA

INT1 0.90 0.34 0.39 0.39 0.38 0.42 0.94 0.34 0.36 0.41 0.35 0.29
INT2 0.86 0.29 0.29 0.38 0.42 0.46 0.95 0.34 0.36 0.40 0.39 0.27
INT3 0.86 0.25 0.28 0.38 0.34 0.30 0.96 0.36 0.33 0.43 0.37 0.28
IMA1 0.37 0.82 0.36 0.36 0.27 0.25 0.31 0.89 0.58 0.57 0.38 0.45
IMA2 0.26 0.81 0.56 0.35 0.33 0.23 0.33 0.93 0.52 0.57 0.40 0.35
IMA3 0.25 0.87 0.46 0.37 0.31 0.18 0.35 0.93 0.52 0.58 0.36 0.35
IMA4 0.23 0.83 0.34 0.43 0.28 0.22 0.34 0.92 0.47 0.57 0.38 0.29
EMA1 0.21 0.41 0.83 0.37 0.30 0.22 0.30 0.49 0.91 0.52 0.44 0.42
EMA2 0.31 0.39 0.86 0.42 0.34 0.23 0.36 0.49 0.88 0.53 0.42 0.37
EMA3 0.32 0.47 0.86 0.42 0.38 0.25 0.35 0.52 0.92 0.57 0.46 0.41
EMA4 0.33 0.38 0.70 0.36 0.25 0.30 0.30 0.53 0.86 0.61 0.49 0.50
SA1 0.36 0.37 0.42 0.83 0.49 0.40 0.39 0.49 0.59 0.84 0.65 0.48
SA2 0.39 0.41 0.45 0.83 0.50 0.39 0.37 0.60 0.62 0.91 0.57 0.46
SA3 0.28 0.27 0.25 0.64 0.35 0.46 0.35 0.53 0.47 0.87 0.49 0.41
SA4 0.35 0.36 0.37 0.82 0.47 0.42 0.42 0.59 0.53 0.92 0.58 0.48
SKA1 0.34 0.29 0.31 0.49 0.86 0.44 0.32 0.34 0.45 0.55 0.91 0.45
SKA2 0.40 0.29 0.30 0.49 0.89 0.40 0.35 0.35 0.46 0.56 0.93 0.48
SKA3 0.38 0.33 0.24 0.49 0.80 0.41 0.40 0.39 0.41 0.60 0.89 0.48
SKA4 0.32 0.23 0.45 0.42 0.70 0.38 0.33 0.40 0.51 0.63 0.87 0.53
TA1 0.34 0.23 0.26 0.45 0.49 0.87 0.29 0.37 0.48 0.49 0.48 0.87
TA2 0.43 0.17 0.28 0.42 0.33 0.75 0.23 0.30 0.40 0.44 0.50 0.89
TA3 0.38 0.24 0.30 0.46 0.40 0.83 0.19 0.33 0.35 0.42 0.42 0.79
TA4 0.32 0.21 0.14 0.37 0.38 0.77 0.27 0.30 0.34 0.35 0.37 0.77

People usually perceive hedonic and intrinsic moti- between perceived accessibility of ICT infrastruc-
vation only when they see that there is some social sup- ture and self-efficacy, and perceived accessibility
port and acceptable quality and satisfaction of hedonic- of online contents and self-efficacy.
related services (Junglas et al., 2013). This explains why Finally, consistent with our conjectures, social access
the relationship between social access and intrinsic had stronger differential impact on intrinsic and extrin-
motivational access was significant, as expected. Social sic motivational access for Korea than for Cambodia
access measures content availability. Respondents were (H5 and H6 were supported). In Korea, where ICT is
able to link specific contents to potential hedonic values advanced, the citizens have high expectations toward
they could obtain as a result of consuming particular content accessibility and affordability. When there is
content. Furthermore, Korea saw much higher impact any lag in getting this information, Koreans will react
of social access on intrinsic motivational access (H6 was more aggressively and this will reduce their motivation
supported). to use ICT.
While technological and social access exerted pos-
itive influences on skills access, there was no differ-
ential impact between Cambodia and Korea (H4 and Implications, limitations, and future
H7 were not supported). Our sample was of univer- research
sity students, who are in an environment where they The digital divide is a critical issue demarcating those
have constant access to ICT. They have probably who ‘have’ from those who ‘have-not’ access to ICT,
taken introductory to ICT courses or have learned placing the latter group at disadvantage. Developed
to use ICT in some of their classes. Therefore, a uni- nations benefit from advanced ICT infrastructure and
versity education system prepares the students in policies that stimulate economic and social growth.
both countries to be on par with each other in ICT Developing nations, on the other hand, are still strug-
learning, thus eliminating any differential impact gling with basic ICT infrastructure. Drawing upon
Chang et al: A three-tier ICT access model for intention to participate online 13

Table 7. Path comparisons.

Path Cambodia Korea

H2: Technological access → Extrinsic Motivational access 0.075 (0.059) N.S. < 0.215 (0.055)***
H5: Social access → Extrinsic Motivational access 0.446 (0.064)*** < 0.570 (0.050)***
H6: Social access → Intrinsic Motivational access 0.436 (0.069)*** < 0.571 (0.052)***
H10: Skills access → Intention 0.324 (0.066)*** > 0.213 (0.055)***
Note: ***: Path coefficient is significant at p <0.001; N.S.: Path coefficient is not significant. One-tailed tests were performed as the
direction of the differences was hypothesized. The number in parentheses is the standard errors.

Table 8. Hypotheses testing results.

Hypothesis Supported Not Supported

Differential impact between Stronger influence for Korea

Korea and Cambodia H2: Technological access → Extrinsic H3: Technological access → Intrinsic
motivational access motivational access
H5: Social access → Extrinsic H4: Technological access → Skills access
motivational access
H6: Social access → Intrinsic H7: Social access → Skills access
motivational access
Stronger influence for Cambodia
H10: Skills access → intention
Same impact for both H1: Technological access → social access
Korea and Cambodia H8: Extrinsic motivational access →
intention
H9: Intrinsic motivational access →
intention

Wixom and Todd (1995), Liaw (2007), and Chang The study also contributes to the field of the digital
et al. (2012), this study developed and empirically divide through its measurement of the four different
tested a 3-tier ICT access model in Korea (a developed types of access and its clarification of the relation-
country) and Cambodia (a developing country). The ships among these access types. This provides
results showed significant differences in the beha- insights for policy makers to devise effective inter-
vioral models of countries with different level of ICT ventions to reduce the digital divide. Since technolo-
development. This study represents an important step gical access is the most fundamental access type,
toward developing theoretical understanding of the governments ought to build good ICT infrastructure
digital divide and responds to calls for increased and make IT hardware and software accessible to all
theoretically-based empirical research on the digital citizens. Once good ICT infrastructure is in place,
divide (Van Dijk and Hacker, 2003; Hsieh et al., 2008). they need to formulate policies and regulations to
More importantly, the observed between-group dif- ensure proper informational and educational contents
ferences give insights into differential behavioral pat- are accessible to the public.
terns between the two countries. When addressing As indicated in the findings, social access exerted
online participation issues, governments, depending stronger effect on motivational access and skills access
on different ICT development index levels, should compared to technological access. As such, to increase
devise strategies and mechanisms that are appropriate citizens’ intention to participate online, ICT policy mak-
to their contexts to stimulate citizens’ participation in ers have to play the role of facilitator to assist companies
the digital society. This approach is applicable even in providing public and private contents that incorporate
within one country, where the conditions of ICT infra- local culture, language, and societal norms. At the same
structure vary across different sections of the country time, they have to consider lowering ICT price baskets
(e.g., between rural and urban areas). to increase people’s motivation to use and learn ICT.
14 Information Development

Figure 5. Path analysis results.

Companies should also charge services at affordable Limitations and future research
prices. Alternatively, they should find other forms of To generalize the results obtained here to other con-
revenues instead of passing the cost to consumers. A texts, we have to be aware of the limitations of the
good example is Google’s strategy in charging adver- research design. First, only university students were
tisement fees from companies rather than end-users. approached to answer the survey. The university envi-
For university students, their motivational access and ronment provides students with the privilege of access
ICT skills access are determined by the availability of to ICT more than any other group in each country.
physical technology and social contents. Therefore, to The issue is more prominent in Cambodia, where
improve their motivations to use ICT and their self con- most citizens cannot afford expensive ICT infrastruc-
fidence in using ICT, policy makers have to provide ture, than in Korea. As shown in differential compar-
basic devices, network connections, and good educa- ison results, even students with privileged access in
tional contents to increase their exposure and learning Cambodia exhibited differences from students in
of ICT. Korea. If sample populations without such privileges
As depicted in the 3-tier ICT access model, were to be surveyed in Cambodia, we might expect
one’s intention to participate online goes beyond even more significant differences. To test this hypoth-
simply having access to basic ICT infrastructure esis, future research should study groups with different
and readily available contents. Instead, policy- demographic characteristics such as age, education,
makers need to instill interest in their citizens to gender, income, and living area (rural versus urban)
encourage their participation. They should alo pro- to understand how they differ in online participation.
vide programs that teach ICT knowledge. Having Future research could also compare two nations with
better knowledge will improve citizens’ confidence similar ICT development index to see how other fac-
toward their own ICT skills sets and increase their tors such as government policies, availability of ICT
intention to participate in online activities.
Chang et al: A three-tier ICT access model for intention to participate online 15

infrastructure, and cultural factors influence the cit- publications/the-world-factbook/geos/cb.html>. Access

izens’ intention to participate online. November, 30, 2013.
This study adopted a cross-sectional design, captur- Chang Y, Shahzeidi M, Kim H and Park MC (2012) Gender
ing only a snapshot of the digital divide phenomenon. digital divide and online participation: A cross-national
Future research could adopt a longitudinal approach analysis. The 19th ITS Biennial International Confer-
ence, Bangkok, Thailand, November 18-21, 2012,
tracing individual changes across a longer time span.
http://www.econstor.eu/dspace/handle/10419/72506.
This may yield richer understanding into how a coun- Chen W and Wellman B (2003) Charting and bridging digital
try’s ICT investment patterns and policies shape its cit- divides: comparing socio-economic, gender, life stage,
izens’ ICT use over time. To overcome the digital and rural-urban Internet access and use in eight countries.
divide, future research could also investigate success- AMD Global Consumer Advisory Board (GSAB).
ful cases of social inclusion and extract useful practices Chin WW (1998) The partial least squares approach for
that can be used to assist those who are deprived of structural equation modeling. In G. A. Marcoulides
ICT. As demonstrated by the results of between- (Ed.), Modern Methods for Business Research (pp.
group differences, these best practices have to be tai- 295–336). Mahwah, NJ: Lawrence Erlbaum Associates.
lored to fit local contexts. Through collaborative effort Chinn M and Fairlie R (2007) The determinants of the global
across different societies and nations, inequalities digital divide: a cross-country analysis of computer and
caused by the digital divide can hopefully be reduced, internet penetration. Oxford Economic Papers 59(1): 16–44.
Compeau DR and Higgins CA (1995) Computer self-
if not totally eradicated.
efficacy: Development of a measure and initial test. MIS
quarterly 19(2): 189–211.
Acknowledgments Cooper J and Weaver KD (2003) Gender and computers:
This research was supported by the MSIP (Ministry of Sci- Understanding the digital divide. Mahwah, NJ: Lawr-
ence, ICT and Future Planning), Korea, under the CPRC ence Erlbaum Associates.
(Communications Policy Research Center) support program Crump B and McIlroy A (2003) The digital divide: Why
supervised by the KCA (Korea Communications Agency) the ‘‘don’t-want-tos’’ won’t compute: Lessons from a
(KCA-2013-001). New Zealand ICT project. First Monday 8(12).
Cuervo MRV and Menendez AJL (2006) A multivariate
References framework for the analysis of the digital divide: Evi-
Agarwal R and Prasad J (1997) The role of innovation char- dence for the European Union-15. Information and
acteristics and perceived voluntariness in the acceptance Management 43(6): 756–766.
of information technologies. Decision Sciences 28(3): Davis FD (1989) Perceived usefulness, perceived ease of
557–582. use, and user acceptance of information technology.
Ajzen I (2002) Perceived behavioral control, self-efficacy, MIS Quarterly 13(3): 319–340.
locus of control, and the theory of planned behavior. Davis FD, Bagozzi RP and Warshaw PR (1989) User accep-
Journal of Applied Social Psychology 32: 665–683. tance of computer technology: a comparison of two the-
Ayanso A, Cho DI and Lertwachara K (2010) The digital oretical models. Management Science 35(8): 982–1003.
divide: global and regional ICT leaders and followers. Davis FD, Bagozzi RP and Warshaw PR (1992) Extrinsic and
Information Technology for Development 16(4): intrinsic motivation to use computers in the workplace1.
304–319. Journal of Applied Social Psychology 22(14): 1111–1132.
Bandura A (1997) Self-efficacy: The exercise of control. DeLone WH and McLean ER (1992) Information systems
New York: W.H. Freeman. success: the quest for the dependent variable. Informa-
Bhattacherjee A (2001) Understanding information sys- tion Systems Research 3(1): 60–95.
tems continuance: An expectation-confirmation model. Dewan S and Riggins FJ (2005) The digital divide: Current
MIS Quarterly 25: 351–370. and future research directions. Journal of the Associa-
Bishop AP, Bazzell I, Mehra B and Smith C (2001) Afya: tion of Information Systems 6(12): 298337.
Social and digital technologies that reach across the DiMaggio P and Hargittai E (2001) From the ‘digital divide’
digital divide. First Monday 6(4). to ‘digital inequality’: Studying Internet use as penetration
Broos A and Roe K (2006) The digital divide in the Play- increases. Princeton University Center for Arts and Cul-
Station generation: Self-efficacy, locus of control and tural Policy Studies, Working Paper Series number, 15.
ICT adoption among adolescents. Poetics 34: 306–317. Eastin MS and LaRose R (2000) Internet self-efficacy and
Castells M (2009) The Network Society. Edward Elgar Pub- the psychology of the digital divide. Journal of
lishing, Northampton, MA. Computer-Mediated Communication 6(1): 0.
CIA (Central Intelligence Agency) (2013) The World Fact Fishbein M and Ajzen I (2005) The influence of attitudes
book. Retrieved from <https://www.cia.gov/library/ on behavior. The Handbook of Attitudes: 173–222.
16 Information Development

Fornell C and Larcker DF (1981) Evaluating structural Kernis MH and Waschull SB (1995) The interactive roles of
equation models with unobservable variables and mea- stability and level of self-esteem: Research and theory. In
surement error. Journal of Marketing Research 18(1): M. P. Zanna (Ed.). Advances in Experimental Social Psy-
39–50. chology (27: 93–141), San Diego, CA: Academic Press.
Gefen D, Karahanna E and Straub DW (2003) Trust and Lenhart A (2002) Barriers to Internet access: From the non-
TAM in on line shopping: An integrated model. MIS user and new user perspective. In Association of Internet
Quarterly 27: 51–90. Researchers. Conference 3.0, Maastricht, Netherlands,
Hair JF, Black WC, Babin BJ and Anderson RE (2009) October 14.
Multivariate Data Analysis. New Jersey: Upper Saddle Liaw SS (2007) Computers and the Internet as a job assisted
River, Pearson Prentice Hall. tool: based on the three-tier use model approach. Com-
Hoffman DL, Novak TP and Schlosser AE (2003) Locus of puters in Human Behavior 23(1): 399–414.
control, Web use, and consumer attitudes toward Inter- Liaw SS (2008) Investigating students’ perceived satis-
net regulation. Journal of Public Policy and Marketing faction, behavioral intention, and effectiveness of
22(1): 41–57. e-learning: A case study of the Blackboard system.
Hsieh JJ, Rai A and Keil M (2008) Understanding digital Computers and Education 51(2): 864–873.
inequality: Comparing continued use behavioral models Liaw SS and Huang HM (2003) An investigation of user
of the socio-economically advantaged and disadvan- attitudes toward search engines as an information
taged. MIS Quarterly 32(1): 97–126. retrieval tool. Computers in Human Behavior 19(6):
Hsieh JJ, Rai A and Keil M (2011) Addressing digital 751–765.
inequality for the socioeconomically disadvantaged Liaw SS, Chang WC, Hung WH and Huang HM (2006)
through government initiatives: Forms of capital that Attitudes toward search engines as a learning assisted
affect ICT utilization. Information Systems Research tool: approach of Liaw and Huang’s research model.
22(2): 233–253. Computers in Human Behavior 22(2): 177–190.
Hsu MH and Chiu CM (2004) Internet self-efficacy and Lin HF (2006) Understanding behavioral intention to par-
electronic service acceptance. Decision Support Systems ticipate in virtual communities. Cyber Psychology and
38(3): 369–381. Behavior 9(5): 540–547.
Huyer S (2005) Women, ICT and the information society: Lin HF (2007) Effects of extrinsic and intrinsic motivation
Global perspectives and initiatives. In Proceedings of on employee knowledge sharing intentions. Journal of
the International Symposium on Women and ICT: Cre- Information Science 33(2): 135–149.
ating Global Transformation, New York: ACM Press. Lin JC and Lu H (2000) Towards an understanding of the
ITU (International Telecommunications Union) (2013) behavioral intention to use a website. International
Measuring the Information Society. Retrieved from < Journal of Information Management 20: 197–208.
http://www.itu.int/en/ITU-D/Statistics/Documents/pub- Ono H and Zavodny M (2007) Digital inequality: A five
lications/mis2013/MIS2013_without_Annex_4.pdf>. country comparison using micro data. Social Science
Access November, 30, 2013. Research 36(3): 1135–1155.
Jackson LA, Ervin KS, Gardner PD and Schmitt N (2001) Parker JS (2001) Searching for the Digital Divide. Informa-
The Racial Digital Divide: Motivational, Affective, and tion Development 17(4): 205–208.
Cognitive Correlates of Internet Use. Journal of Applied Pick JB and Azari R (2008) Global digital divide: Influence
Social Psychology 31(10): 2019–2046. of socioeconomic, governmental, and accessibility fac-
Junglas I, Goel L, Abraham C and Ives B (2013) The Social tors on information technology. Information Technology
Component of Information Systems—How Sociability for Development 14(2): 91–115.
Contributes to Technology Acceptance. Journal of the Selwyn N (2003) Schooling the mobile generation: The future
Association for Information Systems 14(10): 585–616. for schools in the mobile-networked society. British Jour-
Kaba A and Said R (2013) Bridging the digital divide nal of Sociology of Education 24(2): 131–144.
through ICT A comparative study of countries of the Susanto A, Lee H, Zo H and Ciganek AP (2013) Factors
Gulf Cooperation Council, ASEAN and other Arab Affecting Internet Banking Success: A Comparative
countries. Information Development, published online Investigation between Indonesia and South Korea.
9 July 2013, DOI: 10.1177/0266666913489987. Journal of Global Information Management 21(2):
Karahanna E, Straub DW and Chervany NL (1999) Informa- 72–95.
tion technology adoption across time: a cross-sectional Szajna B (1996) Empirical evaluation of the revised tech-
comparison of pre-adoption and post-adoption beliefs. nology acceptance model. Management Science 42(1):
MIS Quarterly 23(2): 183–213. 85–92.
Katz JE and Rice RE (2002) Social consequences of Internet Taylor S and Todd PA (1995). Understanding information
use: Access, involvement and interaction. Cambridge, technology usage: A test of competing models. Informa-
MA: MIT Press. tion Systems Research 6(2): 144–176.
 Chang et al: A three-tier ICT access model for intention to participate online 17

Trinder K, Guiller J, Margaryan A, Littlejohn A and Nicol D Technology and Telecommunications Review as well as
(2008) Learning from digital natives: bridging formal and in the proceedings of international conferences and domes-
informal learning. The Higher Education Academy: 1–71. tic conferences. Contact: Dept. of Business & Technol-
UNDP (United Nations Development Programme) (2013) ogy Management, Korea Advanced Institute of Science
Human Development Report 2013. Retrieved from < and Technology (KAIST), #5108, N22, 291 Daehak-ro,
http://hdr.undp.org/en/media/HDR2013_EN_Summary.
Yuseong-gu, Daejeon, 305-701, Republic of Korea.
pdf>. Access November, 30, 2013.
Tel: þ82423506332. Fax: þ82423506339. E-mail:
Van Deursen A, Van Dijk JA and Peters O (2006) Rethinking
[email protected]
Internet skills: The contribution of gender, age, educa-
tion, Internet experience, and hours online to medium- and
content-related Internet skills. Poetics 39: 125–144. Siew Fan Wong is a senior lecturer in the Department of
Van Dijk JA (2006) Digital divide research, achievements Information Systems at Sunway University, Malaysia. She
and shortcomings. Poetics 34(4): 221–235. received her PhD degree in MIS from the University of
Van Dijk JA and Hacker K (2003) The digital divide as a Houston, Texas. Her research focuses on IT outsourcing
complex and dynamic phenomenon. The Information and IT adoption. Recently, she started working on projects
Society 19(4): 315–326. that look at the impacts of social media usage and digital
Venkatesh V and Brown S (2001) A longitudinal investiga- divide. Her publications have appeared in journals such
tion of personal computers in homes: Adoption determi- as Information Development, Entrue Journal of Informa-
nants and emerging challenges. MIS Quarterly 25(1): tion Technology, and International Journal of Business
71–102. Data Communications and Networking. Contact: Sunway
Vinzi VE, Chin WW, Henseler J and Wang H (2010) Hand-
University - Department of Information Systems, No 5,
book of Partial Least Squares: Concepts, Methods and
Jalan Universiti, Bandar Sunway Petaling Jaya Selangor
Applications. Berlin: Springer-Verlag, Springer Hand-
books of Computational Statistics. 46150, Malaysia. T: 603 7491-8622 F: 603 5635-8633.
Wixom BH and Todd PA (2005) A theoretical integration Email: [email protected], [email protected]
of user satisfaction and technology acceptance. Informa-
tion Systems Research 16(1): 85–102. Myeong-Cheol Park is a professor in the Department of
World Bank (2013) Data: Countries and Economies. Business and Technology Management at Korea Advanced
Retrieved from <http://data.worldbank.org/country>. Institute of Science and Technology (KAIST). He is a board
Access November, 30, 2013. member of CPRsouth (Communication Policy Research
WEF (World Economic Forum) (2013) The Networked South) and former President of Korea Technology Innovation
Readiness Index 2013. Retrieved from <http://www3. Society. He has published numerous papers in the areas of tel-
weforum.org/docs/GITR/2013/GITR_OverallRank- ecommunications management strategy and economics in
ings_2013.pdf>. Access November, 30, 2013.
Telecommunications Policy, Information Economics and
Wu J and Lu X (2013) Effects of Extrinsic and Intrinsic Moti-
Policy, Internet Research, Journal of Global Information
vators on Using Utilitarian, Hedonic, and Dual-Purposed
Management, Scientometrics, Information Technology and
Information Systems: A Meta-Analysis. Journal of the
Association for Information Systems 14(3): 153–191. Management, Information Systems and e-Business Manage-
Zhu S and Chen J (2013) The digital divide in individual ment, Management Decision, and ETRI Journal. One of his
e-commerce utilization in China results from a national papers published in Telecommunications Policy is now
survey. Information Development 29(1): 69–80. honored as the most-cited paper since his publication from
2004. Before joining KAIST, he worked for Electronics and
Telecommunication Research Institute (ETRI), one of the
About the authors
largest Korean government research institutes. He received
Younghoon Chang is a PhD candidate in the Department his BS and MA degrees from Seoul National University, and
of Business and Technology Management at Korea his PhD degree in business administration from the University
Advanced Institute of Science and Technology (KAIST). of Iowa (USA). Contact: Department of Business &
He received his bachelor’s degree in business administra- Technology Management, Korea Advanced Institute of
tion from Korea University. Younghoon’s research inter- Science and Technology (KAIST), #504, N22, 291
ests include digital divide, e-business, ICT strategy, and Daehak-ro, Yuseong-gu, Daejeon, 305-701, Republic
cross-cultural IT marketing. His articles have appeared in of Korea. Tel: þ82423506313. Fax: þ82423506339.
Information Development, Entrue Journal of Information E-mail: [email protected]

View publication stats