In recent years, the Internet-based environment has experienced
prolific development. With multiple forms of representation, some
research has indicated that content delivered in blended or
Internet-based conditions may be more efficient than that is provided in
traditional classrooms (Abdous & Yoshimura, 2010). However, when
comparing the differences between distance education and classroom-based
instruction, some studies have found no significant difference in
effectiveness between distance education within blended or
Internet-based learning settings and face-to-face (F2F) education
(Bernard et al., 2004; Delialioglu & Yildirim, 2008). Indeed, a
variety of results have been derived from the studies relating to
Internet-based learning (IBL). For instance, one of the features of IBL
is that it is supposed to foster learners' active participation in
the construction of knowledge (White & Frederiksen, 2005). Some
studies have indicated positive effects on students' knowledge
construction in IBL processes (Pena-Shaff & Nicholls, 2004), whereas
some have reported that the discussions among learners are generally at
low levels of knowledge construction in IBL systems (Guan, Tsai &
Hwang, 2006). Hence, it may be suggested that due to the divergent
components of the IBL context created by a variety of researchers,
different traits may be produced.
Undoubtedly, a large amount of work has been devoted to
constructing a preferable Internet-based setting to date. When compared
with the traditional classroom setting, IBL provides many appealing
attributes, which may consist of increasing the availability of learning
experiences for learners who cannot or choose not to attend F2F
offerings, assembling and disseminating instructional content more
cost-efficiently, or enabling instructors to handle more students while
maintaining learning outcome quality that is similar to that of
comparable F2F instruction (Abdous & Yoshimura, 2010). Furthermore,
learners can have access to the information without time limits or
location constraints. That is, Internet-based settings may help learning
be unrestricted to any specific moment or to any particular classroom.
Thus, it is generally believed that IBL is likely to provide potential
applications for students' learning activities. However, some
research indicates that users' intention to continue in IBL may be
low (Lee, 2010). In other words, occurrences of participants dropping
out of IBL are not uncommon (Roca, Chiu, & Martinez, 2006). A number
of possible explanations may account for learners' discontinuing
IBL. For instance, more than a decade ago, Katz and Aspeden (1996)
stated that uncertainty about how to get started and the perception that
computers were too complicated were possible
barriers. Indeed, research evidence has indicated that learners are
unwilling to take part in IBL probably because they lack confidence in
operating the system (Eastin & LaRose, 2000).
Based on a similar thought, some researchers have suggested that
students' involvement in IBL may be associated with the perception
of their own capabilities relating to specific skills and knowledge.
They assert that such a concept, often referred to as self-efficacy, may
play an important role in students' learning processes and learning
outcomes in Internet-based classroom settings (Shakpa & Ferrari,
2003). In fact, recent empirical studies (e.g. Hoffman & Spatariu,
2008) have demonstrated that students with higher self-efficacy gain
better performance in contrast to those with lower self-efficacy in
Internet-based settings. Thus, with the significant importance of
self-efficacy in IBL, the aim of this paper is to conduct a literature
review examining the theory, evidence, and application of the
relationship between self-efficacy and IBL. On the basis of such a
concept, the review will firstly define a theoretical framework for
exploring self-efficacy in the Internet-based context. Then, the
evidence of self-efficacy in IBL and how it is connected to the original
concept of self-efficacy will be explained. Finally, a collection of
related empirical studies regarding self-efficacy in the IBL condition
will be reviewed. In this review, three categories regarding the
relations between self-efficacy and IBL are classified:
1) the Internet Self-Efficacy (ISE), which examines learners'
confidence in their general skills or knowledge of operating Internet
functions or applications in the Internet-based learning condition;
2) the interplay between Academic Self-Efficacy and Internet-Based
Learning (ASE&IBL), which investigates the role of learners'
general academic confidence played in the Internet-based learning
3) the Internet-Based Learning Self-Efficacy (IBLSE), which
explores learners' confidence in their participation and their
expected performance, particularly in terms of the Internet-based
In general, self-efficacy refers to how confident an individual
feels about handling particular tasks, challenges, and contexts
(Bandura, 1997). It is widely considered to be derived from
Bandura's (1986) Social Cognitive Theory (SCT). Bandura (1994)
defines self-efficacy as people's beliefs "about their
capabilities to produce designated levels of performance that exercise
influence over events that affect their lives" (p.71). It is
generally reported that individuals with higher self-efficacy perceive
difficult tasks as meaningful challenges, despite the fact that others
may find similar tasks discouraging. In Bandura's (1994)
understanding, high self-efficacy "fosters intrinsic interest and
deep engrossment in activities" (p.71); on the contrary, a lack of
self-efficacy may cause people to have low aspirations, slacken their
efforts, and give up easily. In addition, some researchers (e.g.,
Girasoli & Hannafin, 2008) have further indicated that
learners' cognitive processes can be influenced by self-efficacy.
Furthermore, as proposed by Pajares and Schunk (2001), instead of
being evaluated in general, research regarding self-efficacy should be
assessed at a domain-specific or task-specific level because such
measures may have greater validity and predictive relevance. In other
words, domain-specific self-efficacy assessment, such as asking students
to state their confidence in learning mathematics or writing, is more
explanatory and predictive than omnibus measures and preferable for
making general academic judgments (Pajares, 1996).
Levels of self-efficacy are usually considered to have strong
validity for specific task domains, and most of the findings have
suggested that self-efficacy is positively related to learners'
performance. That is, a strong sense of self-efficacy can enrich human
achievement in many ways (Karsten & Roth, 1998). For example,
Caprara et al. (2008) indicated that the lower the decline in
self-efficacy, the higher the grades and the greater the likelihood of
remaining in high schools. Hoffman and Spatariu (2008) similarly
demonstrated the positive effects of self-efficacy on problem-solving
efficiency. Based on the above literature, when exploring the
relationship between self-efficacy and IBL, it becomes important to
interpret self-efficacy carefully from different perspectives.
Self-efficacy in IBL environment
Recently, a great amount of research relating to self-efficacy has
been carried out in the educational research community. Nevertheless,
different researchers have observed learners' self-efficacy from a
variety of perspectives. As a result, prior to describing relevant
applications of self-efficacy in the IBL research, it may be helpful to
identify appropriate definitions for different types of self-efficacy.
In general, academic self-efficacy (ASE) pertains to a student's
perception of academic learning (Girasoli & Hannafin, 2008), while
computer self-efficacy (CSE) is defined as an individual's
perceived confidence regarding his/her ability to use a computer
(Compeau & Higgins, 1995; Murphy, Coover, & Owen, 1989).
Likewise, general Internet self-efficacy describes people's
perceptions about their own abilities to use the Internet (Tsai &
Tsai, 2003), whereas IBL self-efficacy represents individuals'
confidence and self-belief in their ability to master an online course
or online learning activity (Yukselturk & Bulut, 2007). For the
purpose of this review and for the consistency of terminology,
learners' general Internet self-efficacy is named as their ISE
(Internet self-efficacy) in the present study.
In comparison with the development of computers, Internet
technology is viewed as a relatively innovative invention. Therefore,
before attempting to interpret the conceivable relations between
self-efficacy and IBL, it is meaningful to discuss the relevant findings
concerning CSE. Marakas Yi, and Johnson (1998) defined CSE as "an
individual's perception of efficacy in performing specific computer
related tasks within the general computing domain" (p. 127). Thus,
CSE can be considered a domain specific measure of self-efficacy that
reflects a person's belief in his/her ability to perform specific
computer tasks. Consistent with the original concept of the
self-efficacy theory, CSE is developed over time and is thought to have
influences on the consequence of learners' interactions with
computers when facing obstacles (Compeau & Higgins, 1995; Murphy,
Coover, & Owen, 1989). Compared with ISE-related work, more
extensive literature on CSE has been published. For instance, Moos and
Azevedo (2009) conducted a comprehensive literature review on the
relations between computer-based learning environments and CSE. Based on
their report, a number of scholars have distinguished CSE into disparate
dimensions. For example, Marakas, Yi, and Johnson (1998) and Marakas,
Johnson, and Clay (2007) have divided CSE into two distinct levels:
general computer self-efficacy, which assesses learners' general
beliefs about their computing skills (e.g., their confidence in using
software to complete a computing job), and application-specific
self-efficacy, which assesses confidence in using specific applications
(i.e. confidence in the ability to rename a file in specific
applications such as Excel or Word).
Moos and Azevedo's (2009) work synthesized the studies
including those empirically examined factors related to CSE and the
relationship among CSE, learners' learning outcomes, and learning
processes in the computer-based learning environment. On the basis of
their findings, both behavioral and psychological factors were found to
be positively related to CSE, which is related to students'
learning outcomes in computer-based learning environments. Besides, it
was shown that this relationship may change with students'
acquisition of skills or knowledge. Finally, users' CSE might be
related to their navigational paths in computerized learning
environments. Different from Moos and Azevedo's review on CSE, the
present research specifically focuses on those studies pertaining to the
relations between self-efficacy and IBL environments.
In this study, the Social Science Citation Index (SSCI) database
from 1999 to 2009 was used for paper selection using the following
keywords for topics: Internet AND self-efficacy; web AND self-efficacy;
network AND self-efficacy; e-learning AND self-efficacy; online AND
self-efficacy. The first phase of the search produced 489 articles.
Studies published from 1999 to 2009 were selected because Internet
technology is considered to have been widely implemented in the
educational realm since 1999. To illustrate, in 1999, the UNESCO
Institute on Information Technologies in Education initiated and began
the project the Internet in Education (UNESCO, 2003). Moreover, IBL was
defined as those learning activities taking place in an Internet-based
setting. Then, the data gathering procedure was directed to the
subsequent selection derived from the criteria determined by three
experts in the field of educational technology. The selection criteria
were comprised of three principles: (a) the major purpose of the study
must include at least one component probing the role of self-efficacy in
any kind of IBL condition, (b) the study design should be based on an
empirical methodology, and (c) the main findings of the research must be
related to learning and must elaborate the application of self-efficacy
in an Internet-based setting. Abstracts were first reviewed and articles
were then limited according to these principles. Then, full papers were
examined for the relevancy to this review. On the basis of the
previously mentioned criteria and three rounds of expert panel
discussions for the validation of the selection, 46 articles remained
for the current review.
Four educational researchers examined the 46 papers selected,
conducted content analyses by summarizing the major findings of the
studies, and after two rounds of discussions, concluded three categories
for this review, which could cover almost all of the topics under
investigation. The first category consisted of the studies relating to
learners' general Internet self-efficacy (ISE); the second category
included the investigation exploring the interplay between
learners' academic self-efficacy and the Internet-based learning
(ASE&IBL). The third category contained research probing
learners' IBL self-efficacy (IBLSE), that is, learners' self
confidence in their participation and their expected performance in an
Moreover, similar to the framework applied in other reviews (e.g.
Lee et al., in press; Tallent-Runnels et al., 2006), two subcategories
were further drawn. The first subcategory comprised the studies
exploring the relationship between students' self-efficacy and
their learning process or learning outcomes in IBL conditions.
Meanwhile, the second subcategory was made up of the research probing
how students' self-efficacy might be altered among different IBL
contexts. Table 1 provides an outline of the research framework of the
Internet Self-efficacy (ISE)
According to Table 1, studies in the ISE category were divided into
two subcategories. On the one hand, the investigation between
learners' ISE and their learning processes or outcomes in the IBL
condition was explored. On the other hand, an amount of research was
utilized to probe how learners' ISE may have altered in IBL
settings. A complete list of research involving learners' ISE is
summarized in Table 2.
Relations between ISE and learning processes/outcomes
It was found that some studies had been conducted to assess
learners' basic perceptions of ISE in IBL prior to the
investigation of the relationship between learners' ISE and their
learning processes or outcomes in the IBL condition. For instance,
Torkzadeh and van Dyke (2001), Wu and Tsai (2006), and Peng, Tsai, and
Wu (2006) developed a number of questionnaires in order to assess
learners' basic perceptions of ISE. Torkzadeh and van Dyke (2001)
used 277 responses from university students to develop and validate a
17-item ISE scale. Statistical analysis supported a three-factor model,
including surfing/browsing, encryption/decryption, and system
manipulation. According to their report, the first factor assessed
learners' confidence in surfing, browsing or finding information in
an IBL setting; the second factor assessed learners' confidence in
decrypting or encrypting messages in an Internet-based setting; the
third factor assessed learners' confidence in operating an IBL
system. Evidence of reliability and construct validity were indicated in
Similarly, to find out learners' fundamental perceptions of
ISE, Tsai and his colleagues (Peng, Tsai, & Wu, 2006; Wu & Tsai,
2006) divided ISE into two types: general Internet self-efficacy and
communicative Internet self-efficacy. General self-efficacy addresses
students' Internet self-efficacy for basic functions or purposes
(e.g. I can search for information on the Internet by using keywords),
whereas communicative self-efficacy probes their efficacy for
Internet-based communication or interaction (e.g. I think I can talk to
others in online chat rooms). With a sample of 1,313 university students
in Taiwan, Wu and Tsai (2006) found that students' Internet
attitudes are highly correlated with not only general ISE but also with
communicative ISE. It was suggested that students' Internet
attitudes could be viewed as one of the important indicators for
predicting ISE. In a similar way, Peng, Tsai, and Wu (2006) investigated
1,417 Taiwanese university students and proposed that students
perceiving the Internet as a leisure tool (e.g. as a tour or a toy)
showed higher communicative ISE than those using the Internet simply as
In addition to probing learners' basic perceptions of ISE,
Tsai and his research team also attempted to identify the conceivable
relationship between learners' ISE and their learning processes or
learning outcomes (Chu & Tsai, 2009; Liang & Tsai, 2008; Tsai
& Tsai, 2003). For example, to examine the role of students'
ISE in their information searching strategies in an IBL setting, Tsai
and Tsai (2003) conducted 8 in-depth case studies and concluded that
high ISE students had better information searching strategies and
learned better than those with low ISE in the Internet-based condition.
Moreover, in an attempt to explore the relationship between
learners' ISE and their preferences for IBL, Liang and Tsai (2008)
surveyed 365 Taiwanese college students and revealed that students with
higher ISE (e.g. "I can search for information on the Internet by
using keywords.") showed greater preferences for IBL which they
could use with ease; however, students with higher communicative ISE
(e.g. "I think I can talk to others in online chat rooms.")
tended to display relatively weaker preferences for inquiry learning in
Recently, to establish a theoretical model to explain factors that
might influence adult learners' preferences for constructivist IBL
settings, Chu and Tsai (2009) gathered data from 541 participants
enrolled in adult education institutes for structural equation modeling
(SEM) analyses. The results revealed that ISE plays a mediating role in
the relationships between Internet usage and the participants'
preference for IBL, and indicates that with augmented time spent on
adult learners' ISE, strengthening their preferences for IBL,
may be increased.
Besides exploring how ISE might be associated with learner's
learning processes, a certain amount of research has investigated the
relationship between ISE and the subsequent use of Internet-based
systems. For instance, Yi and Hwang (2003) extended the technology
acceptance model by incorporating ISE to predict the use of the
Blackboard system by surveying 109 university students, and concluded
that ISE positively influences the decision to use Internet-based
technology and subsequent actual use. Contradictory to Yi and
Hwang's positive results, in an attempt to explore factors
influencing students' likelihood of using the Internet to seek
information, Lu et al. (2007) found that respondents' ISE had no
significant association with their intentions to seek information on the
Internet after surveying 229 international students. Similarly, with a
sample of 368 undergraduates, Yang et al. (2007) found that the anxiety
of Internet use negatively influenced ISE, whereas ISE did not
significantly affect the intention to use Internet sites. Lam and Lee
(2006) inquired into the role of ISE and outcome expectations in
learners' usage of the Internet by a longitudinal study among 1,000
seniors in Hong Kong. Their findings generally validated the effects of
ISE and outcome expectations on the Internet usage intention.
Finally, Hong (2006) explored the effects of ISE and search task
specificity on the outcomes and task perseverance of finding online
health-related sites that contained attributes of website
accountability. In the study, 84 US university students conducted two
search tasks (general and specific) that varied in the degree of task
difficulty. The results showed that high ISE participants located sites
higher in website accountability in the general search task than their
low ISE counterparts. Besides, the participants with high ISE
demonstrated more task perseverance than those with low ISE.
Alteration of ISE
A number of studies were found to utilize experimental designs and
to examine how learners' ISE might be changed through an
intervention or training. Take Schmidt and Ford's (2003) study for
example, 42 undergraduate students received a brief introduction to
metacognitive practices, in which trainees were informed to more
frequently and accurately reflect on what they were learning through the
program before creating Web pages, while 37 participants in the control
condition began the Web-page creation training immediately. Consistent
with their expectations, learners reporting greater levels of
metacognitive activity during training had higher levels of ISE when
compared with their counterparts.
Similarly, O'Malley and Kelleher (2002) required 55 university
students majoring in public relations to develop a statement and
measured their ISE before, immediately after, and 7 weeks after working
in either geographically dispersed (Kansas and Hawaii) or local (Kansas
only) teams. In the experimental section, two participants from Kansas
State and two from Hawaii were randomly assigned to each group. The
participants in the control section were also randomly assigned, but
they were not assigned to collaborate with students from Hawaii. The
results revealed that learners' ISE increased over time regardless
of the experimental conditions.
Torkzadeh and his team members conducted several experimental
studies to probe how ISE might be elevated or related to learners'
Internet attitudes. Torkzadeh and van Dyke (2002) as well as Torkzadeh,
Chang, and Demirhan (2006) reported the effects of training on
learners' ISE and their computer user attitudes. With a 17-item ISE
scale developed and validated in 2001, Torkzadeh and van Dyke (2002)
reported on the effects of training on students' ISE and their
computer user attitudes by utilizing questionnaires with a sample of 189
university students. Training was considered an important way of
improving computer-related self-efficacy (Compeau & Higgins, 1995;
Marakas, Yi, & Johnson, 1998). Therefore, the study collected
questionnaire responses at both the beginning and end of an introductory
computer training course. The content of the training program was not
clearly reported in Torkzadeh and van Dyke's (2002) study, but they
suggested that the training significantly improved the students'
ISE. Besides, respondents with a favorable attitude toward computers
were found to have higher ISE scores than those with an unfavorable
attitude; and male respondents consistently reported higher than females
for ISE on both the pre- and post-training scores.
Torkzadeh, Chang, and Demirhan (2006) developed and examined a
contingency model of learners' CSE and ISE. With measures of user
attitude, computer anxiety, computer self-efficacy, and Internet
self-efficacy, the authors analyzed the survey responses of 347
university students from multiple sections of a training course like
information technology infrastructure and decision support systems.. The
result suggested that the training programs significantly improved
learners' CSE and their ISE. Besides, respondents with favorable
attitudes toward computers improved their ISE significantly more than
those with unfavorable attitudes. Respondents with low computer anxiety
improved both their CSE and ISE significantly more than those with high
computer anxiety; however, the interaction effect between attitude and
anxiety was only significant for the CSE scores but not for the ISE
Finally, Chiou and Wan (2007) investigated the change of ISE on
information searching in an Internet-based condition. The students
receiving low-difficulty manipulation (i.e., allowing a longer search
period) obtained a higher level of ISE, whereas those receiving
high-difficulty manipulation (i.e., allowing a shorter search period)
possessed a lower level of ISE. The results indicated that the
enhancement effect of positive task experience (such as low-difficulty
tasks) on self-efficacy was more pronounced for individuals with lower
levels of ISE on information searching in Internet-based settings,
whereas the deteriorating effect of negative experience was more
prominent for individuals with higher levels of ISE on information
searching in the Internet-based condition.
Summary of ISE research
In conclusion, research found in the ISE category mainly focused on
developing and validating methods of assessing learners' ISE and
exploring its relationship with those factors likely to play a role in
students' learning processes or outcomes in the IBL condition. To
name a few, the relations among students' ISE and their attitudes,
strategies, and preferences were examined.
Besides, some researchers also paid attention to gender-related
issues while making efforts to link ISE with the abovementioned
constructs. It is generally believed that computer-related tasks are
more advantageous for males than females (Li & Kirkup, 2007).
Although some studies (e.g. Wu & Tsai, 2006; Torkzadeh & van
Dyke, 2002) have actually found that male students reveal better ISE
than their female counterparts, a report of 234 high school participants
conducted by Brown et al. (2003) suggested that either boys' or
girls' ISE regarding a specific simulation ILE named GlobalEd
Project had revealed similar patterns.
Interestingly, when taking measures to explore learners'
levels of ISE, a majority of the studies (e.g. Joo, Bong, & Choi,
2000; Thompson, Meriac, & Cope, 2002; Tsai & Tsai, 2003) adopted
the behavior or the performance of search tasks as students'
learning outcomes. To illustrate, Thompson, Meriac, and Cope (2002)
found positive correlation between ISE and the number of correct search
results produced. Joo, Bong, and Choi (2000) stated that students'
scores on the Internet-based search tests were significantly and
positively predicted by their ISE. Hence, it may be suggested that
search tasks is regarded as the most commonly implemented IBL activities
at the present stage.
Interplay between academic self-efficacy and Internet-based
According to Bandura (1997), ASE is defined as students'
expectations of how successful they will be in the classroom. There is
no doubt that issues relating to ASE have been extensively researched
(Pajares, 1996; Schunk & Pajares, 2002); however, most of the
investigations are not related to the learning occurring in
Internet-based contexts. Therefore, in the present review, the
ASE&IBL category was specifically drawn to discuss the interplay
between learners' ASE and IBL settings.
At first, it was found that a variety of methods and objectives
characterized the research comprising the ASE&IBL category. To name
a few, ASE was disclosed to be associated with goal orientation (Sins et
al., 2008), self-regulated learning (Crippen & Earl, 2007;
Yukselturk & Bulut, 2007; Joo et al., 2000), and motivational
beliefs (Yukselturk & Bulut, 2007; Tai, 2006) in IBL settings.
Moreover, it was found that, on the one hand, most of the
non-experimental studies have investigated the relationship between
learners' ASE and their motivational constructs influencing
pupils' IBL processes or outcomes. On the other hand, the
experimental research was inclined to the comparisons among diverse
types of learning environments. More specifically, to probe potential
variations, some were conducted within a variety of forms of the IBL
setting, but some were implemented within both Internet-based and
traditional F2F learning conditions. It was proposed that different
extents of students' ASE may have been derived from the differences
in these learning environments. The studies involving ASE&IBL issues
are summarized in Table 3.
Relations between ASE and IBL processes/outcomes
As stated, a number of studies (i.e. Sins et al., 2008; Waldman,
2003; Yukselturk & Bulut, 2007) have explored how ASE, coupled with
other motivational constructs, plays a role in students' successful
IBL. That is, they have sought to find out how ASE may be associated
with students' IBL processes or IBL outcomes.
Sins et al. (2008) tested 60 11th-grade students' conceptual
model of the relationship between students' achievement goal
orientation and their ASE with respect to a modeling task in an
Internet-based setting, where learners could collaborate online by means
of a synchronous chat on inquiry assignments for science courses. The
study found that learners' mastery-approach goal orientation and
their ASE were both positively related to their achievement in the
modeling task and their use of deep cognitive processes.
Similarly, in an attempt to find out what might influence
students' IBL of programming, Yukselturk and Bulut (2007) analyzed
and examined the relationship among 80 online learners' selected
variables (i.e. gender, age, educational level, locus of control, and
learning style), motivational beliefs (i.e. intrinsic goal orientation,
extrinsic goal orientation, control beliefs, task values, self-efficacy,
and test anxiety), self-regulated learning components, and their success
in an Internet-based setting. The study result showed that although
learners' ASE and their intrinsic goal orientation beliefs were
correlated with their IBL success, they did not enter the final
prediction model in the regression analyses. This finding was somewhat
contradictory to Pintrich and de Groot's (1990) research outcomes,
in which respondents' ASE and their intrinsic motivation
significantly affected their achievement.
In a similar way, to encourage students to use library facilities
and electronic resources, Waldman (2003) conducted a study to understand
what factors may have promoted students to seek out information in a
library setting. Previous research showed that students' ASE might
be related to their academic achievement outcomes (Bandura, 1997;
Pajares, 1996). Therefore, Waldman (2003) analyzed 340 university
freshmen's responses concerning their library/computer usage and
their ASE. The research finding showed that the students who expressed
interest in learning about the library's electronic resources were
more likely to have higher ASE for completing the learning task.
Moreover, students with higher ASE for completing the task tended to use
the library more often. The study outcome was consistent with other
research on self-efficacy (e.g. Ren, 2000), suggesting that
self-efficacious students have a tendency to be more active in academic
work and to use resources available to them.
Finally, Thompson, Meriac, and Cope (2002) examined the
relationship between learners' self-efficacy (including both
general ISE and ASE) and their search task performance in the IBL
condition. A total of 90 participants were required to search the
Internet and to list the names of the industrial-organizational
psychologists they found. The findings indicated that the improvement in
both ISE and ASE could lead to higher online performance.
Alteration of ASE by IBL
Whereas the research discussed above mainly deals with how ASE may
influence or be related to students' learning processes and
outcomes in an Internet-based setting, the following studies observe how
ASE may be altered in different Internet-related learning conditions. It
was found that some studies measured the learners' ASE within IBL
conditions only (e.g. Crippen & Earl, 2007; Meyer et al., 2002),
while others intended to compare learners' differences of ASE
between traditional F2F classroom settings and Internet-based situations
(e.g. Francescato et al., 2006, 2007; Kitsantas & Chow, 2007).
Crippen and Earl (2007), Meyer et al. (2002), and Tai (2006) have
evaluated participants' ASE among Internet-based conditions only.
Crippen and Earl (2007) described a quasi-experimental study, wherein
expert modeling was believed to improve ASE, and worked examples served
as expert models in their study. A total of 66 students were randomly
assigned to one of three conditions: a worked example group, a worked
example/self-explanation group, and the control group. In the end, the
combination of worked example with self-explanation prompt was reported
to improve students' performance, problem solving skills, and ASE
in terms of whether personal goals were achieved.
Meyer et al. (2002) assessed the impact of using a structured
strategy as a base for an intergenerational Internet tutoring program,
in which 12 older adults provided tutoring for 5th-grade students to
learn the strategy through an instructional Internet-based system. The
structured strategy was considered to allow readers to build mental
representations similar to the text's hierarchical organization of
important ideas. Sixty students were randomly assigned to one of three
groups: (a) a tutoring group, in which the students worked on the
Internet-based system using the structured strategy with a tutor; (b) a
group in which the students worked independently on the same
Internet-based instruction without a tutor; and (c) a control group, in
which the students did not receive instruction in the structured
strategy. The results showed that both tutors and children in the
structured strategy group with tutors increased their ASE.
Likewise, to examine the effects of training framing from
supervisors on trainees' ASE and training motivation in IBL, Tai
(2006) surveyed 126 employees entering a training program introducing
computer software operation and design, and further tested how these
variables may have influenced the overall training effectiveness. The
126 employees were asked to complete a series of questionnaires at the
beginning, the midpoint, and the end of the course. The results
indicated that supervisors training framing could be used to predict
trainees' ASE., which subsequently affected their reactions,
learning, and motivation.
As mentioned earlier, other researchers have intended to identify
the differences in learners' ASE between traditional F2F classrooms
and IBL settings. The exploration of such differences may illustrate the
role or the effects of IBL on students' ASE. For example,
Francescato et al (2006, 2007) compared learners' self-efficacy
between traditional F2F and IBL conditions within a computer-supported
collaborative learning (CSCL) setting in particular. Francescato et al.
(2006) implemented a pilot study, in which 50 psychology major students
were required to learn the same material in the F2F and Internet-based
classroom settings. The results indicated that participants in both
groups achieved similar growth in their levels of ASE, social
self-efficacy, and self efficacy for problem solving. Collecting data
from a different sample, Francescato et al. (2007) conducted the other
study with 166 students in similar experimental conditions. Different
from the previous study, the results of the second research found
statistically significant increases only in learners' social
self-efficacy and self-efficacy for problem solving for both groups, but
not in their ASE. On the basis of their study outcomes, no significant
increase in learners' ASE for both F2F and Internet-based CSCL
conditions was found; thus, it may be suggested that the Internet-based
CSCL environment could be regarded as efficient as traditional F2F
classroom settings in increasing learners' social self-efficacy and
self-efficacy for problem solving, but the effects on ASE may vary
across the two studies.
Kitsantas and Chow (2007) examined how college students'
help-seeking behavior varied across three different instructional
learning environments. A total of 472 students enrolled in distance,
distributed, and traditional classes were queried about their
help-seeking preferences, help-seeking tendencies, personal threat in
seeking help, and ASE. The research findings showed that, regardless of
the class in which they were enrolled, the students' academic
achievement was positively associated with their ASE. These results were
consistent with previous research findings, in which ASE was positively
inter-correlated and predicted achievement, and students with higher ASE
for successful problem solving displayed greater performance monitoring
and persisted longer than those with lower ASE (Pintrich & de Groot,
Finally, in an attempt to examine the changes in self-efficacy
through Internet-based courses, one study concerning medical
professional training was reported. Farel, Umble, and Polhamus (2001)
discussed the effect of an analytical skills training course on medical
professional development and practice. Through a one-year Internet-based
program, the study found that 28 participants' ASE increased
significantly, suggesting that the Internet-based analytical and
technical training initiatives could offer a promising means for
reaching public health professionals, and provide an alternative
opportunity for off-site workshops. With the IBL, in-service
professional practitioners could acquire easier access to and adoption
of training to meet their needs, which may have led to greater
motivation as well as increased ASE.
Summary of ASE&IBL research
Because students' perceptions of ASE are considered to be
important in their use of self-regulated strategies (Zimmerman &
Martinez-Pons, 1990), an amount of research (e.g. Crippen & Earl,
2007; Yukselturk & Bulut, 2007; Joo, Bong, & Choi, 2000) has
probed the potential interplay between learners' ASE and their
self-regulated learning activities in Internet-based settings.
Furthermore, a number of studies (e.g. Joo, Bong, & Choi, 2000;
Brown, et al., 2003; Thompson, Meriac, & Cope, 2002) were found to
query not only learners' ASE but also their ISE. For instance, Joo,
Bong, and Choi (2000) tested the applicability of self-efficacy theory
to the contexts of a specific Internet-based condition, in which
learners had to conduct several search tasks. They found that
learners' perceptions of ASE could predict their performance
measured by written tests, whereas their perceptions of ISE were
significant in predicting their search test performance. These results
give evidence to support that learners' ASE is more associated with
achievement measured by a conventional assessment mode, while
learners' ISE is related to their performance in operating
Internet-related functions. Similarly, Brown et al. (2003) discussed
gender issues in terms of learners' ASE and ISE. They found no
difference between the two genders and concluded that both boys and
girls revealed similar patterns of responses for both kinds of
self-efficacy. Finally, Thompson, Meriac, and Cope (2002) examined
learners' ASE and ISE in the IBL condition and suggested that
improvement in both self-efficacies could lead to higher online
performance. This conclusion was consistent with Bandura's (1986)
assertion, indicating a reciprocal interaction between learners'
self-efficacy and their performance. Therefore, when Girasoli and
Hannafin (2008) reviewed the potential importance of designing scaffolds
in the Internet-based condition, they suggested that both students'
ASE and their general ISE should be taken into consideration and
Internet-based learning self-efficacy (IBLSE)
The category of IBLSE is made up of the research which examines
learners' confidence in their participation and their expected
consequent performance particularly derived from IBL activities. A
number of features were found among the studies.
First of all, rather than developing a full instrument particularly
evaluating learners' IBLSE, most researchers only included a factor
with a limited number of questions in their surveys. Several reasons may
account for a lack of relevant research. First, a large number of
participants are required for the development of surveys, and it may
somehow be difficult for researchers to have access to a large group of
individuals possessing complete IBL experiences. IBLSE-related
instruments can be context dependent and susceptible to a specific kind
of IBL programs. Thus, it becomes difficult for other researchers to
validate a questionnaire previously utilized in other research with
different types of IBL activities. Thus, IBLSE-related instrument or
factor may only play a minor role in many of the studies (e.g. Chang
& Tung, 2008; Liaw, 2008; Tsai, 2009).
Besides, compared with the research concerning ISE or ASE&IBL,
studies relating to IBLSE seem to be widely applied to certain
established models. A certain amount of research was found to comply
with the Technology Acceptance Model (TAM) and its associated
implementation regarding the IBLSE construct.
Finally, very little research aimed to investigate the changes of
learners' IBLSE. Instead of implementing experimental designs, a
large amount of research has merely presented relational/co-relational
data within this category (e.g. Artino, 2008; Johnson, Hornik, &
Salas, 2008; Wang & Newlin, 2002). A limited number of experimental
studies may suggest that, rather than observing the possible differences
among diverse learning conditions, IBLSE-related research seems to have
explored the interplay between learners' self-efficacy in terms of
IBL and their academic outcomes in such learning activities, or their
satisfaction with Internet-based programs. A complete list of studies
involving IBLSE issues is provided in Table 4.
Relations between IBLSE and IBL processes/outcomes
As stated, most of the findings revealed in the IBLSE-related
studies pertain to certain established models. Consequently, IBLSE is
frequently utilized as a predicting factor which may influence or be
related to students' learning processes or outcomes. Moreover, it
was also noted that a majority of the research pays attention to the
application of the Technology Acceptance Model (TAM). For example, Park
(2009) surveyed 628 university students' adoption of IBL using the
structural equation modeling technique. A general structural model
including IBLSE was developed, and IBLSE was found to be the most
important construct for the participants' intention to use
e-learning. Similarly, Lee (2006) probed the factors affecting the
adoption of an IBL system by surveying 1,085 Taiwanese university
students online. The research evidence was found to lend support to the
original TAM findings, in which students' IBLSE was demonstrated to
significantly relate to their perceived ease of use. Hence, it was
considered important to develop an easy-to-use system and to increase
In a similar way, on the basis of questionnaire responses collected
from 67 female and 89 male employees in six international companies
based in Taiwan, Ong and Lai (2006) explored gender differences in
perceptions and relationships among factors affecting the acceptance of
IBL. The research findings showed that although females' rating of
IBLSE was lower than males', females' perception of IBLSE
played a more important determinant role in affecting their behavioral
intention to take part in IBL. It was hence suggested that gender issues
should be taken into consideration when developing relevant theories.
Furthermore, in a series of studies probing learners'
intention to engage in IBL, Chang and Tung (2008) and Tung and Chang
(2008a, 2008b) proposed new hybrid models in which TAM was combined with
the innovation diffusion theory. According to their research results,
IBLSE was one of the critical factors which may have an impact upon
students' behavioral intentions. At first, Chang and Tung (2008)
combined the innovation diffusion theory and TAM, and added two research
variables comprising of perceived system quality and IBLSE to study
students' behavioral intentions to use an IBL course. With an
analysis of questionnaire responses from 212 undergraduate students who
were using online learning course websites in Taiwan, the study found
that IBLSE had a positive effect on students' behavioral intention
to use the online learning course websites.
Similarly, derived from the questionnaire responses from 267
nursing students of six universities in Taiwan, Tung and Chang (2008a)
reported that IBLSE had a positive effect on learners' behavioral
intention to use the Internetbased nursing program. Finally, Tung and
Chang (2008b) added four variables, including computer anxiety, IBLSE,
perceived financial cost, and perceived information quality. Based on
228 questionnaires collected from nursing students who had taken
Internet-based courses in Taiwan, they found the more confident students
were in their ability to use IBL (i.e., higher IBLSE), the more likely
they were to take part in Internet-based courses.
Likewise, deriving from the notion of TAM, a three-tiered
Technology Use Model (3-TUM) was developed by Liaw, Huang, and Chen
(2007) and Liaw (2008) to investigate how individuals' IBLSE
influences their satisfaction and behavioral intention regarding IBL
programs. In their studies, 3-TUM was defined as integrated
multidisciplinary perspectives comprising motivation, social cognitive
theory, theory of planned behavior, and TAM. In Liaw, Huang, and
Chen's (2007) study, a total of 30 instructors were asked to answer
a series of questionnaires. The result indicated that instructors'
behavioral intention to use the IBL program was positively influenced by
their perceived IBLSE. Similarly, derived from 424 university
students' survey responses, Liaw (2008) stated that users'
perceived IBLSE played a positive role in determining students'
satisfaction with and behavioral intention to use Internet-based
Though Lee and Lee (2008) did not directly refer their study to the
notion of TAM, they proposed a research model adopting IBLSE as a
moderating variable to investigate learners' perceptions of the
quality of an IBL system. Their study result indicated that higher IBLSE
group was more sensitive to the effectiveness and usefulness of the
system and was aware of the contextual information quality (e.g. the
variety of the lectures) than those with lower IBLSE. On the contrary,
the lower IBLSE group was found to be more sensitive to the effect of
the ease-of-use of the system and paid more attention to the
representational information quality (e.g. the consistency of the
lectures). This outcome may suggest that learners with different IBLSE
have diverse opinions about IBL activities.
While some studies have specialized in issues concerning TAM,
others are concerned with the relationship between learners'
perceived IBLSE and their satisfaction with IBL settings. As an
instance, Artino (2008) surveyed 646 undergraduates and concluded that
learners' IBLSE and their perception of the learning environment
were significantly positive predictors of their satisfaction. Likewise,
Johnson, Hornik, and Salas (2008) developed a model of e-learning
effectiveness, which added social presence to other frequently studied
variables, such as users' IBLSE, perceived usefulness, course
interaction, and effectiveness. With an examination of survey responses
from 345 individuals, they found that learners with higher IBLSE were
more satisfied with the course than those with lower IBLSE.
Still others demonstrated the relationship between students'
IBLSE and their learning outcomes in the IBL condition. For instance, to
examine students' personal choices when taking Internet-based
courses, Wang and Newlin (2002) investigated 122 college students and
tested whether learners' IBLSE would predict their performance in
the Internet-based sections of the class. The results showed that
students' perceived IBLSE were predictive of their final exam
scores. Moreover, students showing curiosity about the Internet-based
program revealed higher IBLSE and had better class performance than
those taking part in the course solely due to availability.
Similarly, Bolman et al. (2007) investigated the usability of a
navigation support tool, which guided learners by generating advice on
the next best step to take in a self-directed Internet-based course.
Although they found that the navigation tool had not increased
learners' IBLSE, it was indicated that learners with high IBLSE had
completed more modules, adhered more often to the advice given, and were
convinced that the navigation tool helped them plan the course.
Therefore, it was suggested to incorporate IBLSE enhancing strategies in
the navigational support of IBL activities.
Alteration of IBLSE
Among the reviewed papers, little research directly examined how
IBLSE might be altered by certain types of IBL. Rather, these studies
utilized "indirect" methods of investigation to reveal some
potential avenues of fostering IBLSE. For instance, Bates and Khasawneh
(2007) considered that evaluating the mediating role of IBLSE could
provide a better understanding of the functional properties or potential
enhancement of IBLSE and further clarify what factors might account for
the differences among individuals in their participation in IBL
activities. Accordingly, they proposed a mediated model to seek and to
identify a number of theoretically based factors, which were believed to
contribute to the development of IBLSE. On the basis of 288 university
students' survey responses and selfreports, the research results
revealed a partially mediated model, in which the block of antecedents
(i.e. students' previous success with the IBL, instructor feedback,
anxiety, pre-course training, and the perceived nature of IBL ability)
had a direct effect on the dependent variables (i.e. students'
outcome expectations, mastery perceptions, and the hours spent per week
using the IBL technology to complete assignments for university courses)
as well as an indirect effect through their influence on IBLSE. The
finding was considered consistent with Bandura's (1982) premise
that one of the strongest sources of self-efficacy beliefs is an
individual's direct experience with the same or a similar
phenomenon. It was proposed that instructional strategies, providing
positive learning experiences with the IBL, may play a vital role in
enhancing learners' IBLSE, fostering positive expectations, and
encouraging their use of the technology.
With a sample of 223 learners taking part in an Internet-based
program, Choi, Kim, and Kim (2007) confirmed that flow experience and
attitude towards IBL had significant impacts on learners' IBLSE.
Therefore, to enhance the effectiveness of IBL, it may not be sufficient
to focus solely on learners' preferences; instead, to increase
students' experience involvement, or intrinsic interest may be some
possible ways to enhance learners' preferences, which may
consequently contribute to students' IBLSE.
Besides, in an attempt to investigate students' conceptions of
learning in the F2F condition, conceptions of IBL, and the differences
between these conceptions, Tsai (2009) analyzed 83 Taiwanese college
students' interview transcripts. The findings derived several
categories of conceptions of F2F traditional-type learning and IBL, and
it was suggested that the conceptions of IBL were often more
sophisticated than those of F2F learning. In addition, learners'
questionnaire responses revealed that the sophistication of the
conceptions of IBL was associated with better searching strategies as
well as higher IBLSE. These findings highlighted the need for fostering
students' conceptions of learning by Internet-based environment, as
they may enhance more sophisticated learning strategies and IBLSE.
Finally, different from previous research designs, one study was
found to compare students' learning outcomes within a variety of
classroom settings. Moneta and Kekkonen-Moneta (2007) assessed 414
students on not only affective learning (including intrinsic engagement,
extrinsic engagement, and negative affect) but also IBLSE in an
introductory computing course, which was taught once in a lecture format
and twice in a rich interactive multimedia online format. IBLSE was
assessed by a questionnaire item. The research results found that the
IBL modules fostered more intrinsic engagement and higher IBLSE.
Summary of IBLSE research
In conclusion, while looking into the research relating to
learners' IBLSE, probing their confidence in the participation and
expected performance in the Internet-based activities, a great number of
studies (e.g. Chang & Tung, 2008; Liaw, Huang, & Chen, 2007)
were found to deal with the extended development of established models.
For example, the notion of TAM is extensively applied in the relevant
research. Moreover, some studies (e.g. Artino, 2008; Johnson, Hornik,
& Salas, 2008) have discussed the relation between users' IBLSE
and their satisfaction with IBL, whereas others (e.g. Bates &
Khasawneh, 2007; Choi, Kim, & Kim, 2007; Tsai, 2009) were found to
perceive IBLSE as a predictor or a mediator of students' learning
outcomes in an Internet-based setting.
Indeed, by assessing students' IBLSE, researchers may have
acquired indications about their expected outcomes derived from IBL
activities. In fact, this finding may have also resulted from the
fundamental feature of IBLSE, in which IBLSE was utilized to evaluate
learners' perceptions of their learning in an Internet-based
setting. Therefore, it may have been somehow unavoidable to associate
IBLSE with the students' evaluation of their satisfaction or
performance with regards to IBL.
Compared with the research found in the ISE and ASE&IBL
categories, it was noted that relatively few studies (Moneta &
Kekkonen-Moneta, 2007) had been conducted to compare learners'
IBLSE among different learning environments. However, it was found that
providing positive learning experiences with IBL activities (Bates &
Khasawneh, 2007; Choi, Kim, & Kim, 2007) or the sophistication of
the conceptions of IBL conceptions of IBL (Tsai, 2009) may have played
an important role in enhancing learners' IBLSE.
Finally, several studies were found to probe learners' IBLES
in specific domains in particular. For example, Chang and Tung (2008)
and Tung and Chang (2008a, 2008b) examined students' IBLES in
nursing contexts, whereas others investigated their IBLES in IBL of
management (Lee & Lee, 2008), psychology (Wang & Newlin, 2002)
or service academy (Artino, 2008). In general, the research results have
indicated a positive influences of IBLES on either learners'
intention, performance, or satisfactory toward IBL.
Discussions and Conclusions
State of self-efficacy research in IBL environments
Because of the increasingly important role self-efficacy plays in
Internet-based learning (IBL), the relationship between self-efficacy
and IBL has been widely investigated in the last decade. Consequently,
the present study has collected and investigated 46 research papers
published from 2000 to 2009 concerning these relations for a
comprehensive literature review. Research is classified into three major
categories: the Internet Self-Efficacy (ISE); the interplay between
Academic Self-Efficacy and Internet-Based Learning (ASE&IBL); and
the Internet-Based Learning Self-Efficacy (IBLSE).
Regarding the category of ISE research, the studies generally focus
on the relationship between learners' Internet self-efficacy and
learning processes. The relations among students' ISE and their
attitudes, strategies, and preferences have been frequently examined. It
is worth noting that because of the prolific development of
Internet-based instruction, which has elicited various forms of IBL
activities, research concerning ISE is likely to assess students'
confidence in their skills or knowledge of operating specific Internet
applications (such as communication) in IBL contexts instead of
evaluating learners' ISE in general.
As for research regarding ASE&IBL, it is found that
students' ASE is often applied to correlate with their performance,
motivation, and perceptions of the effectiveness of Internet-based
systems. In general, students' ASE has had positive effects on
their academic outcomes resulted from IBL.
Finally, research on learners' IBLSE, which investigates
learners' confidence in their expected performance in IBL, is
discovered to mainly deal with the application of established models.
For instance, a great amount of research has been arranged on the notion
of Technology Acceptance Model (TAM). Moreover, IBLSE was found to be
perceived as a predictor of students' learning outcomes and their
satisfaction with IBL activities. In general, IBLSE was shown to have
impacts on learners' satisfaction with IBL.
Evaluation of self-efficacy research in IBL environments
According to the original theory proposed by Bandura (1982, 1994),
the source of self-efficacy is derived from multiple sources of efficacy
information, including enactive mastery (e.g., past performance
accomplishments resulting from previous experiences or training), verbal
persuasion such as that resulting from collaboration and
performance-related corrective feedback, and physiological arousal
including changes in emotional states such as anxiety, fear, or positive
anticipation. However, except for Francescato et al. (2006, 2007) and
Johnson, Hornik, and Salas' (2008) reports concerning social
persuasion, Bates and Khasawneh (2007), Yang, et al. (2008), and Chiou
and Wan's (2007) investigation on mastery experiences, and Moneta
and Kekkonen-Moneta's (2007) paper regarding affection arousal,
relatively few empirical studies were found researching from the
initially proposed concept of selfefficacy.
In addition, deriving from Moos and Azevedo's (2009) review of
computer self-efficacy (CSE), three major findings were suggested:
First, both learners' behavioral and psychological factors are
related to CSE; secondly, CSE pertains to students' learning
outcomes in computer-based learning (CBL) environments; and finally, CSE
was found to be associated with users' navigational paths. In
comparison with their findings, the current study found similar trends
but somehow varied results in the research on self-efficacy in IBL
environments. For instance, the results of research concerning the
relationship between students' self-efficacy and their behavioral
factors were found to be inconsistent in this current study. On one
hand, some study findings revealed that ISE (Chu & Tsai, 2009; Lam
& Lee, 2006; Yi & Hwang, 2003) or IBLSE (Chang & Tung, 2008;
Liaw, Huang, & Chen, 2007) are related to students' subsequent
use of IBL systems; on the other hand, others (Lu et al., 2007; Yang et
al., 2007) stated that ISE had no significant effect on their following
participation in IBL activities. This might imply there is a difference
between the roles of CSE and ISE played in CBL and IBL environments
respectively, indicating different natures could exist between CSE and
ISE as well as between CBL and IBL.
Aside from the inconsistent findings about the relations between
learners' self-efficacy and their behaviors, investigation outcomes
on the relationship between students' self-efficacy and their
psychological factors (such as perceived attitude, anxiety, and
usefulness) seem to be consistent in this current study. Studies have,
in general, indicated a positive relation between students' ISE and
their attitude towards IBL (Peng, Tsai, & Wu, 2006; Torkzadeh,
Chang, & Demirhan, 2006; Wu & Tsai, 2006) and a negative
relation between learners' ISE and their perceived anxiety (Lam
& Lee, 2006; Torkzadeh, Chang, & Demirhan, 2006; Yang et al.,
Moreover, consistent with Moos and Azevedo's (2009) findings,
the relationship between students' self-efficacy and their achieved
outcomes in IBL was found to be positively correlated. For instance,
Tsai and Tsai (2003) stated that students with higher ISE had better
information search strategies and learnt better than their counterparts.
Thompson, Meriac, and Cope (2002) found positive relations between
students' self-efficacy (including both ISE and ASE) and the number
of correct search results produced. In the ASE&IBL category, various
researchers (Sins et al., 2008; Wang & Newlin, 2002; Yukselturk
& Bulut, 2007) have claimed that ASE could serve as a positive
predictor of learners' final achievement in the IBL condition.
Similarly, in the IBLSE category, Bolman et al. (2007) also revealed a
positive relationship between learners' IBLSE and their IBL
Furthermore, it was noted that when attempting to take measures
probing learners' perceptions of self-efficacy, a significant
amount of research (e.g. Chiou & Wan, 2007; Joo, Bong, & Choi,
2000; Thompson, Meriac, & Cope, 2002) has adopted search tasks to
predict students' learning outcomes in the Internet-based setting
because search tasks may still be considered as the most commonly
implemented IBL activities. Nevertheless, in contrast to Moos and
Azevedo's (2009) findings concerning CSE, no relevant research in
this review was found on the relationship between individuals'
perceived self-efficacy in the Internet-based setting and their
navigational paths. More research may be needed to investigate on the
Finally, some methodological issues may be worthy of notice. First,
it seems that all of the Internet-related research concerning
self-efficacy is based on questionnaires or surveys for measuring
self-efficacy. Researchers should find other ways of assessing
students' Internet-related self-efficacy, such as interviews or
observation. Most of the studies in this review employed a quantitative
approach; qualitative or mixed research approaches are recommended for
future research. In addition, among the 46 papers reviewed, 35 studies
used university students as their samples. In other words, most of the
participants invited to take part in the related studies were either
undergraduates or graduate students in universities. It may be necessary
to encourage learners with various kinds of demographic backgrounds to
take part in the relevant research. Meanwhile, among the reviewed
papers, 19 studies in America, 22 studies in Asia, and 5 studies in
Europe were reported. Relevant research based on European samples is
relatively rare. Finally, within all of the reviewed studies, only
students' or employees' perceptions were probed; it may be
interesting to investigate instructors' perceptions of
Because the papers selected for the current review were limited to
those included in the SSCI database, other relevant research regarding
self-efficacy may still be found to outline a more comprehensive review.
Future studies can explore the differences between students'
perceptions about CBL and IBL as well as compare students' CSE and
ISE, simultaneously, in order to further examine the relationship
between CSE and ISE. Secondly, researchers can further examine the
construct of three categories of self-efficacy in IBL by assessing the
significance and power of using ISE, ASE and IBLSE to predict
students' IBL performances or outcomes. This could help researchers
and educators realize more about the relationships among the three
variables and learning outcomes in IBL environments. Thirdly, future
studies can examine the relationship between students' perceived
self-efficacy and their learning behaviors in specific Internet-based
learning context such as the online search tasks mentioned above.
Finally, more qualitative methods are suggested for future
Internet-related self-efficacy assessments.
Funding for this research work is supported by the National Science
Council, Taiwan, under grant numbers NSC 992511-S-011-005-MY3 and
(papers with * are those selected for current review)
* Artino, A. R. (2008). Motivational beliefs and perceptions of
instructional quality: Predicting satisfaction with online training.
Journal of Computer Assisted Learning, 24, 260-270.
Abdous, M., & Yoshimura, M. (2010). Learner outcomes and
satisfaction: A comparison of live video-streamed instruction, satellite
broadcast instruction, and face-to-face instruction. Computers &
Education, 55, 733-741.
Bandura, A. (1982). Self-efficacy mechanism in human agency.
American Psychologist, 37, 122-147.
Bandura, A. (1986). Social foundations of thought and action.
Englewood Cliffs, NJ. Prentice Hall.
Bandura, A. (1994). Self-efficacy. In V. S. Ramachaudran (Ed.),
Encyclopedia of human behavior (Vol. 4, pp. 71-81). New York: Academic
Bandura, A. (1997). Self-efficacy: The exercise of control. New
York: W.H. Freeman and Company.
* Bates, R., & Khasawneh, S. (2007). Self-efficacy and college
students' perceptions and use of online learning systems. Computers
in Human Behavior, 23, 175-191.
Bernard, R. M., Abrami, P. C., Lou, Y., Borokhovski, E., Wade, A.,
Wozney, L., Wallet, P.A., Fiset, M., & Huang, B. (2004).
How does distance education compare with classroom instruction? A
meta-analysis of the empirical literature. Review of Educational
Research, 74, 379-439.
* Bolman, C., Tattersall, W., Waterink, J., Janssen, B., van der
Berg, R., van Es, R., & Koper, R. (2007). Learners' evaluation
of a navigation support tool in distance education. Journal of Computer
Assisted Learning, 23, 384-392.
* Brown, S. W., Boyer, M. A., Mayall, H. J., Johnson, P. R., Meng,
L., Butler, M. J., Weir, K., Florea, N., Hernandez, M. & Reis, S.
(2003). The GlobalEd project: gender differences in a problem-based
learning environment of international negotiations. Instructional
Science, 34, 255-276.
Caprara, G. V., Fida, R., Vecchione, M., Del Bove, G., Vecchio, G.
M., Barbaranelli, C., & Bandura, A. (2008). Longitudinal analysis of
the role of perceived efficacy for self-regulated learning in academic
continuance and achievement. Journal of Educational Psychology, 100,
* Chang, S.C., & Tung, F. C. (2008). An empirical investigation
of students' behavioural intentions to use the online learning
course websites. British Journal of Educational Technology, 39, 71-83.
* Chiou, W.-B. & Wan, C.-S. (2007). The dynamic change of
self-efficacy in information searching on the Internet: Influence of
valence of experience and prior self-efficacy. The Journal of
Psychology, 141, 589-603.
* Choi, D. H., Kim, J., & Kim, S. H. (2006). ERP training with
a web-based electronic learning system: The flow theory perspective.
International Journal of Human-Computer Studies, 65, 223-243.
* Chu, R. J.-C., & Tsai, C.-C. (2009). Self-directed learning
readiness, Internet self-efficacy and preferences towards constructivist
Internet-based learning environments among higher-aged adults. Journal
of Computer Assisted Learning, 25, 489-501.
Compeau, D. R., & Higgins, C. A. (1995). Computer
self-efficacy: development of a measure and initial test. MIS Quarterly,
* Crippen K. J. & Earl, B. L., (2007). The impact of Web-based
worked examples and self-explanation on performance, problem solving,
and self-efficacy. Computers & Education, 49, 809-821.
Delialioglu, O., & Yildirim, Z. (2008). Design and development
of a technology enhanced hybrid instruction based on MOLTA model: its
effectiveness in comparison to traditional instruction. Computers &
Education, 51, 474-183.
Eastin, M. A., & LaRose, R. L. (2000). Internet self-efficacy
and the psychology of the digital divide. Journal of Computer Mediated
Communication, 6. Available at:
http://jcmc.indiana.edu/vol6/issue1/eastin.html (October 23, 2010)
* Farel, A,, Umble, K., & Polhamus, B. (2001). Impact of an
online analytic skills course. Evaluation & the Health Professions,
* Francescato, D., Mebane, M.E., Porcelli, R., Attanasio C., &
Pulino M. (2007). Developing professional skills and social capital
through computer supported collaborative learning in university
contexts. International Journal of Human-Computer Studies, 65, 140-152.
* Francescato, D., Porcelli, R., Mebane, M., Cudetta, M., Klobas,
J., & Renzi, P. (2006). Evaluation of the efficacy of collaborative
learning in face to face and computer supported university contexts.
Computers in Human Behavior, 22, 163-176.
Girasoli, A. J., & Hannafin, R. D. (2008). Using asynchronous
AV communication tools to increase academic self-efficacy. Computers
& Education, 51, 1676-1682.
Guan, Y.-H., Tsai, C.-C., & Hwang, F.-K. (2006). Content
analysis of online discussion on a senior-high-school discussion forum
of a virtual physics laboratory. Instructional Science, 34, 279-311.
Hoffman, B., & Spatariu, A. (2008). The influence of
self-efficacy and metacognitive prompting on math problem-solving
efficiency. Contemporary Educational Psychology, 33, 875-893.
* Hong, T. (2006). The Internet and tobacco cessation: The roles of
Internet self-efficacy and search task on the information-seeking
process. Journal of Computer-Mediated Communication, 11, 536-556.
* Johnson, R. D., Hornik, S. & Salas, E. (2008). An empirical
examination of factors contributing to the creation of successful
e-learning environments. International Journal of Human-Computer
Studies, 66, 356-369.
* Joo, Y., Bong, M., & Choi, H.-J. (2000). Self-efficacy for
self-regulated learning, academic self-efficacy, and Internet
selfefficacy in Web-based instruction. Educational Technology Research
and Development, 48, 5-17.
Katz, J. E., & Aspeden, P. (1996, October). Motivations for and
barriers to Internet usage: Results of a national public opinion survey.
Paper presented at the 24th Annual Telecommunications Policy Research
Conference, Solomons, MD.
Karsten, R., & Roth, R. (1998). Computer self-efficacy: A
practical indicator of student computer competency in introductory IS
courses. Informing Science, 3, 61-68.
* Kitsantas, A., & Chow, A. (2007). College students'
perceived threat and preference for seeking help in traditional,
distributed, and distance learning environments. Computers &
Education, 48, 383-395.
* Lam, J. C. I., & Lee, M. K. O. (2006). Digital
inclusiveness-longitudinal study of Internet adoption by older adults.
Journal of Management Information Systems, 22, 4, 177-206.
* Lee, J. K. & Lee, W. K. (2008). The relationship of
e-Learner's self-regulatory efficacy and perception of e-Learning
environmental quality. Computers in Human Behavior, 24, 32-47.
Lee, S. W.-Y., Tsai, C.-C., Wu, Y-T., Tsai, M.-J., Liu, T.C.,
Hwang, F-K., Lai, C.H., Liang, J.-C., Wu, H-C., & Chang, C.-Y. (in
press). Internet-based science learning: a review of journal
publications. International Journal of Science Education.
* Lee, Y.-C. (2006). An empirical investigation into factors
influencing the adoption of an e-learning system. Information &
Management, 30, 517-541.
Lee, M.-C. (2010). Explaining and predicting users'
continuance intention toward e-learning: An extension of the
expectation-confirmation model. Computers & Education, 54, 506-516.
Li, N., & Kirkup, G. (2007). Gender and cultural differences in
Internet use: A study of China and the UK. Computers & Education,
* Liang, J.-C. & Tsai, C.-C. (2008). Internet self-efficacy and
preferences toward constructivist Internet-based learning environments:
A study of pre-school teachers in Taiwan. Educational Technology and
Society, 11, 226-237.
* Liaw, S.-S. (2008). Investigating students' perceived
satisfaction, behavioral intention, and effectiveness of e-learning: A
case study of the Blackboard system. Computers & Education, 51,
* Liaw, S.-S., Huang, H.-M., & Chen, G.-D. (2007). Surveying
instructor and learner attitudes toward e-learning. Computers &
Education, 49, 1066-1080.
* Lu, H. Y., Case, D. O., Lustria, M. L. A., Kwon, N., Andrews, J.
E., Cavendish, S., & Floyd, B. R. (2007). Predictors of online
information seeking by international students when disaster strikes
their countries. CyberPsychology & Behavior, 10, 709-712.
Marakas, G. M., Johnson, R. D., & Clay, P. F. (2007). The
evolving nature of computer self-efficacy construct: An empirical
investigation of measurement construction, validity, reliability and
stability over time. Journal of the Association for Information Systems,
Marakas, G. M., Yi, M. Y., & Johnson, R. D. (1998). The
multilevel and multifaceted character of computer self- efficacy: Toward
clarification of the construct and an integrative framework for
research. Information Systems Research, 9, 126-163.
* Meyer, B. J. F., Middlemiss, W., Theodorou, E., Brezinski, K. L.,
McDougall, J., & Bartlett, B. J. (2002). Effects of structure
strategy instruction delivered to fifth-grade children via the Internet
with and without the aid of older adult tutors. Journal of Educational
Psychology, 94, 486-519.
* Moneta, G. B., & Kekkonen-Moneta, S. S. (2007). Affective
learning in online multimedia and lecture versions of an introductory
computing course. Educational Psychology, 27, 51-74.
Moos, D. C., & Azevedo, R. (2009). Learning with computer-based
learning environments: a literature review of computer self-efficacy.
Review of Educational Research, 79, 576-600.
Murphy, C.A., Coover, D., & Owen, S.V. (1989). Development and
validation of the computer self-efficacy scale. Educational and
Psychological Measurement, 49, 893-899.
* Ong, C.-S. & Lai, J.-Y. (2006). Gender differences in
perceptions and relationships among dominants of e-learning acceptance.
Computers in Human Behavior, 22, 816-829.
* O'Malley, M., & Kelleher, T. (2002). Papayas and
pedagogy: Geographically dispersed teams and Internet self-efficacy.
Public Relations Review, 28, 175-184.
Pajares, F. (1996). Self-efficacy beliefs in achievement settings.
Review of Educational Research, 66, 543-578.
Pajares, F. & Schunk, D. H. (2001). Self-beliefs and school
success: self-efficacy, self-concept, and school achievement. In R.
Riding & S. Rayner (Eds.), Perception (pp. 239-266). London: Ablex
* Park, S. Y. (2009). An analysis of the technology acceptance
model in understanding university students' behavioral intention to
use e-learning. Educational Technology & Society, 12, 150-162.
Pena-Shaff, J. B., & Nicholls, C. (2004). Analyzing student
interactions and meaning construction in computer bullet in board
discussions. Computers & Education, 42, 243-265.
* Peng, H., Tsai, C.-C., & Wu, Y.-T. (2006). University
students' self-efficacy and their attitudes toward the Internet:
The role of students' perceptions of the Internet. Educational
Studies, 32, 73-86.
Pintrich, P. R., & de Groot, E. V. (1990). Motivational and
self-regulated learning components of classroom academic performance.
Journal of Educational Psychology, 82, 33-40.
Ren, W-H (2000). Library instruction and college student
self-sufficiency in electronic information searching. Journal of
Academic Librarianship, 26, 323-328.
Roca, J. C., Chiu, C.-M., & Martinez, F. J. (2006).
Understanding e-learning continuance intention: An extension of the
technology acceptance model. International Journal of Human-Computer
Studies, 64, 683-696.
* Schmidt, A. M. & Ford, J. K. (2003). Learning within a
learner control training environment: The interactive effects of goal
orientation and metacognitive instruction on learning outcomes.
Personnel Psychology, 56, 405-429.
Schunk, D. H., & Pajares, F. (2002). The development of
academic self-efficacy. In A. Wigfield & J. S. Eccles (Eds.),
Development of achievement motivation (pp. 15-31). New York: Academic
* Sins, P. H. M., van Joolingen, W. R., Savelsbergh, E. R., &
van Hout-Wolters, B. (2008). Motivation and performance within a
collaborative computer-based modeling task: Relations between
students' achievement goal orientation, self-efficacy, cognitive
processing, and achievement. Contemporary Education Psychology 33,
* Tai, W. T. (2006). Effects of training framing, general
self-efficacy and training motivation on trainees' training
effectiveness, Personnel Review, 35, 51-65.
Tallent-Runnels, M. K., Thomas, J. A., Lan, W. Y., Cooper, S.,
Ahern, T. C., Shaw, S. M., & Liu, X. (2006). Teaching courses
online: A review of the research. Review of Educational Research, 76,
* Thompson, L. F., Meriac, J. P., & Cope, J. G. (2002).
Motivating Online Performance: the influences of goal setting and
Internet self efficacy. Social Science Computer Review, 20, 149-160.
* Torkzadeh, G., Chang, J.-C., and Demirhan, D. (2006). A
Contingency Model of Computer and Internet Self-efficacy. Information
and Management, 43, 541-550.
* Torkzadeh, G., & Van Dyke, T. P. (2001). Development and
validation of an Internet self-efficacy scale. Behaviour and Information
Technology, 20, 275-280.
* Torkzadeh, G., & Van Dyke, T. P. (2002). Effects of training
on Internet self-efficacy and computer user attitudes. Computers in
Human Behavior, 18, 479-494.
* Tsai, C.-C. (2009). Conceptions of learning versus conceptions of
web-based learning: The differences revealed by college students.
Computers & Education, 53, 1092-1103.
* Tsai, M.-J., & Tsai, C.-C. (2003). Information searching
strategies in web-based science learning: the role of Internet
selfefficacy. Innovations in Education and Teaching International, 40,
* Tung, F.-C. & Chang, S.-C. (2008a). A new hybrid model for
exploring the adoption of online nursing courses. Nurse Education Today,
* Tung, F.-C. & Chang, S.-C. (2008b). Nursing students'
behavioral intention to use online courses: A questionnaire survey.
International Journal of Nursing Studies, 45, 1299-1309
UNESCO (2003). Internet in education: Support material for
educators. Available at: <
http://www.iite.ru/pics/publications/en/files/3214612.pdf> (Oct 30,
* Waldman, M. (2003). Freshmen's use of library electronic
resources and self-efficacy. Information Research, 8, 1-27.
* Wang, A. Y., & Newlin, M. H. (2002). Predictors of
Web-Student Performance: the role of self- efficacy and reasons for
taking an on-line class. Computers in Human Behavior 18, 151-163.
White, B., & Frederiksen, J. (2005). A theoretical framework
and approach for fostering metacognitive development. Educational
Psychologist, 40, 211-223.
* Wu, Y.-T., & Tsai, C.-C. (2006). University students'
Internet attitudes and Internet self-efficacy: A study at three
universities in Taiwan. CyberPsychology & Behavior, 9, 441-450.
* Yang, H.-J., Lay, Y.-L., Tsao, W.-Y., Liou, Y.-C., & Lin,
C.-K. (2007). The impact of language anxiety and self-efficacy on
accessing Internet sites, CyberPsychology and Behavior, 10, 226-233.
* Yi, M. Y., & Hwang, Y. (2003). Predicting the use of
web-based information systems: self-efficacy, enjoyment, learning goal
orientation, and the technology acceptance model, International Journal
of Human-Computer Studies, 59, 431-449.
* Yukselturk, E. & Bulut, S. (2007). Predictors for student
success in an online course. Educational Technology & Society, 10,
Zimmerman, B. J., & Martinez-Pons, M. (1990). Student
differences in self-regulated learning: Relating grade, sex, and
giftedness to self-efficacy and strategy use. Journal of Educational
Psychology, 82, 51-59.
Chin-Chung Tsai (1), Shih-Chyueh Chuang (2), Jyh-Chong Liang (2)
and Meng-Jung Tsai (1)
(1) Graduate Institute of Digital Learning and Education, National
Taiwan University of Science and Technology, Taiwan // (2) Graduate
Institute of Applied Science and Technology, National Taiwan University
of Science and Technology, Taiwan // email@example.com //
firstname.lastname@example.org // email@example.com //
Table 1. The review framework for the research regarding self-
efficacy and Internet-based learning
Category Subcatesory 1 Subcategory 2
Role of self- Alteration of self-
efficacy in IBL efficacy in IBL
ISE (Internet Self- To investigate To probe how
Efficacy) relations between learners' ISE may be
learners' ISE and altered in IBL
learning processes or
outcomes in IBL
ASE&IBL (Academic To investigate the To probe how
Self-Efficacy and interplay between learners' ASE may be
Internet-Based learners' ASE and altered in IBL
Learning) learning processes or
outcomes in IBL
IBLSE (Internet- To investigate To probe how
Based Learning Self- relations between learners' IBLSE may
Efficacy) learners' IBLSE and be altered in IBL
learning processes or
outcomes in IBL
Table 2. Summary of ISE research in alphabetic order. (* = included in
both ISE and ASE&IBL categories; exp=experimental design)
* Brown et al. Find ISE changes through GlobalEd Project
Chiou & Wan To investigate changes of ISE on information
(2007) searching in the Internet-based condition
Chu & Tsai To build a model explaining ISE's influence on
(2009) adult learners' preferences for IBL
Hong (2006) To assess ISE's role in health-related online
* Joo et al. (2000) To test the applicability of self-efficacy
theory to the context of IBL
Lam & Lee (2006) To investigate the role of ISE and outcome
expectations in older adults' usage of
Liang & Tsai (2008) To explore the relations between ISE and
preferences for constructivist IBL
Lu et al. (2007) To explore ISE's effect on students'
likelihood of using Internet-based systems to
O'Malley & To examine PR students' ISE in either
Kelleher(2002) geographically dispersed or local teams
Peng et al. (2006) To explore ISE & perceptions of Internet
Schmidt & Ford To evaluate the impact of meta-cognition
(2003) activities on ISE in IBL
* Thompson et al. To examine the relationship between learners'
(2002) self-efficacy and their search task
Torkzadeh & Van To develop an appropriate ISE instrument
Torkzadeh & Van To examine the relationship among users'
Dyke (2002) training, attitude, and ISE
Torkzadeh et al. To develop and examine a contingency model of
Tsai & Tsai (2003) To examine ISE's effect on information search
strategies for IBL science learning
Wu et al. (2006) To investigate the relationship between
students' attitude and their ISE
Yang et al. (2007) Explore how ISE mediates IAN & IBL intention
Yi & Hwang (2003) To predict the use of IBL by incorporating ISE
Author Participants Method
* Brown et al. 234 high school students in US survey
Chiou & Wan 136 college students in Taiwan exp
Chu & Tsai 541 adult learners in Taiwan survey
Hong (2006) 84 US university student exp
* Joo et al. (2000) 152 junior high school students in survey
Lam & Lee (2006) 1000 adults in Hong Kong exp
Liang & Tsai (2008) 365 college students in Taiwan survey
Lu et al. (2007) 229 university students in US survey
O'Malley & 55 university students in US quasi-exp
Peng et al. (2006) 1417, university, Taiwan survey
Schmidt & Ford 79 undergraduates in US exp
* Thompson et al. 90 undergraduates in US exp
Torkzadeh & Van 277 undergraduates in US survey
Torkzadeh & Van 189 university students in US survey
Torkzadeh et al. 347 university students in US survey
Tsai & Tsai (2003) 8 university freshmen in case
Wu et al. (2006) 1313 university students in Taiwan survey
Yang et al. (2007) 368, university, Taiwan survey
Yi & Hwang (2003) 109, university, US survey
Table 3. Summary of ASE&IBL research in alphabetic order. (* =
included in both ISE and ASE&IBL categories; exp=experimental design)
* Brown et al. To find ISE changes through GlobalEd Project
Crippen & Earl To create an Internet-based program which
(2007) improves user performance and supports well-
structured problem solving
Farel et al. (2001) To show IBL effect on professional development
Francescato et al. To compare learners' ASE between collaborative
(2006) F2F and IBL groups
Francescato et al. To compare learners' self-efficacy in
(2007) developing professional skills in
collaborative F2F and IBL courses
* Joo et al. (2000) To test the applicability of self-efficacy
theory to contexts of IBL
Kitsantas & Chow To examine how ASE varied across four
(2007) different instructional environments
Meyer et al. (2002) To assess the impact of using the Internet-
based setting where elders provided tutoring
Sins et al. (2008) To test relations among learners' AS, goal
orientation, cognitive processing, and
achievement in collaborative IBL
Tai (2006) To test how the effect of training framing
from supervisors on trainees' AS may influence
the overall effectiveness
* Thompson et al. To examine the relationship between learners'
(2002) self-efficacy and their search task
Waldman (2003) To examine the role AS played in students' use
of the library's electronic resources
Yukselturk & Bulut To examine relationship among learner selected
(2007) variables, AS, self-regulated learning, and
their success in IBL
Author Participants Method
* Brown et al. 234 high school survey
(2003) students in US
Crippen & Earl 66 university quasi-exp
(2007) students in US
Farel et al. (2001) 28 staffs in US exp
Francescato et al. 50 university exp
(2006) students in Italy
Francescato et al. 166 university survey
(2007) students in Italy interview
* Joo et al. (2000) 152 junior high school students survey
Kitsantas & Chow 472 college quasi-exp
(2007) students in US
Meyer et al. (2002) 12 adults (62-80 yrs old) and 60 exp
5th-graders in US
Sins et al. (2008) 60 pre-university science majors survey
Tai (2006) 126 employees in Taiwan exp
* Thompson et al. 90 undergraduate exp
(2002) students in US
Waldman (2003) 340 university students in US survey
Yukselturk & Bulut 80 online students survey
(2007) in Turkey interview
Table 4. Summary of IBLSE research in alphabetic order.
Artino (2008) To investigate the relations between learners'
IBLSE and their satisfaction
Bates & Khasawneh To propose a mediated model where a set of
(2007) antecedent variables influenced students'
Bolman et al. To investigate the usability of the IBL system
Chang & Tung To add IBLSE as one of the factors to propose
(2008) a new hybrid TAM.
Choi et al. (2007) To suggest an IBL success model based on flow
Johnson et al. To develop a model by adding social presence
(2008) to learners' IBLSE.
Lee (2006) To investigate factors affecting the adoption
of the IBL through TAM
Lee & Lee (2008) Suggest a research model based on relations of
IBLSE and IBL system perception
Liaw et al. (2007). To explore instructors' and learners'
attitudes toward IBL
Liaw (2008) To examine relations among learner
satisfaction, IBLSE, and IBL effectiveness
Moneta et al. To evaluate students' affective learning in
Ong & Lai (2006). To explore gender differences among dominants
affecting IBL acceptance
Park (2009) To investigate students' adoption of IBL via
SEM technique with LISREL program
Tsai (2009) To examine differences between conceptions of
learning and of IBL
Tung & Chang To study nursing students' behavioral
(2008a) intentions to use IBL
Tung & Chang To propose a new hybrid technology acceptance
(2008b) model (TAM)
Wang & Newlin To investigate if students' IBLSE would
(2002) predict their performance
Author Participants Method
Artino (2008) 646 undergraduates in US survey
Bates & Khasawneh 288 university students survey
(2007) in US
Bolman et al. 808 unders, friends & exp
(2007) families in Holland
Chang & Tung 212 undergraduates in survey
Choi et al. (2007) 223 vocational school students quasi-exp
Johnson et al. 345 university students survey
(2008) in US
Lee (2006) 1085 university students in survey
Lee & Lee (2008) 225 unders in Korea survey
Liaw et al. (2007). 30 instructors & 168 unders in survey
Liaw (2008) 424 university students in survey
Moneta et al. 414 undergraduates in quasi-exp
(2007) Hong Kong
Ong & Lai (2006). 156 employees in Taiwan survey
Park (2009) 628 university students in Korea survey
Tsai (2009) 83 college students in Taiwan interview
Tung & Chang 267 university students in survey
Tung & Chang 228 university students in survey
Wang & Newlin 122 college students in US quasi-exp