312 International Journal of Electronic Business Management, Vol. 8, No. 4, pp.

312-322 (2010)

THE EFFECTIVENESS OF E-LEARNING FOR BLENDED

COURSES IN COLLEGES:
A MULTI-LEVEL EMPIRICAL STUDY
Wenchieh Wu1* and Lan-Yin Hwang2
1
Department of Business Administration
National Taipei University
Taipei (10478), Taiwan
2
Department of Business Administration
National Chengchi University
Taipei (11605), Taiwan

ABSTRACT
Teaching and learning are no longer restricted to traditional classrooms, while e-learning
(electronic learning) has become one of the powerful supporting tools which have
diversified the traditional context of learning in colleges. With the rapid development of
technology, the Internet as a delivery platform has motivated colleges to invest their
resources on developing online programs. Meanwhile, the blended course, which combines
online components with the conventional face-to-face components, has emerged as
alternative mode of teaching and learning and a substantial supplement. With opportunities
and barriers as well, however, the development and management of e-learning are still
challenging, especially for the continuous improvement of students’ learning effectiveness
via e-learning for blended courses. This paper applies the socio-technical systems theory to
review and integrate theories on students’ e-learning from a macro view. To make up the
insufficiency of related research, literature review is conducted first, and an analysis model
is constructed to thoroughly explore factors affecting e-learning effectiveness. Later,
through a questionnaire survey on students’ adoption of e-learning and subsequent
multi-level data analysis, hypotheses on the relationship of the influencing factors and the
research model are verified. Results show that e-learning effectiveness (usefulness of
e-learning, use, and e-learning performance) is simultaneously or alternately affected by
direct or moderating factors of the technical system and the social system at the learning
environment level and the individual level. Compared with the existing research, this paper
uses a more comprehensive system view to construct the theoretical model and empirically
verify it. The results can be a reference for future researchers and managers of e-learning in
colleges.

Keywords: e-Learning, Socio-technical System, Multi-level Analysis

1. INTRODUCTION materials and consistent information. The learning

* content is easy to update. With the rapid development
As e-learning is emerging as the new paradigm of technology, the Internet as a delivery platform has
of modern education, the blended course, which motivated colleges to invest their resources on
combines online components with the conventional developing online programs.
face-to-face components, has emerged as alternative Researchers from various fields have been
mode of teaching and learning and a substantial trying to evaluate the effectiveness of e-learning.
supplement. E-learning has become one of the However, some of them are focused on
powerful supporting tools which have diversified the technology-based components of e-learning system
traditional context of learning in colleges. On one [17] and some are focused on human factor of
hand, e-learning is not confined to geographical e-learning system [27]. These individual assessment
barriers. Students can engage in self-directed learning, frameworks comply with the needs only partially.
and learning resources can be repeatedly used. On the Students’ e-learning system consists of many
other hand, e-learning provides flexible learning subsystems, such as personal factors, technical
environment, and social environment. Without a
multi-level analysis, it is not possible to grasp a
*
Corresponding author: sebwooo@gmail.com comprehensive view of applications of e-learning in
W. C. Wu and L. Y. Hwang: The Effectiveness of e-Learning for Blended Courses in Colleges 313

college/university from the system perspective. After Following the study of Kim [22], we use
reviewing literature associated with e-learning, Ozkan “usability” to estimate users’ attitudes, encompassing
and Koseler [33] proposed that systematic and users’ perceived satisfaction, effectiveness, and
multi-level consideration of evaluation of e-learning efficiency towards the e-learning system. Hence,
systems is necessary and considering the social issues, usability, the attitude toward e-learning system, can
the socio-technical systems approach adopted in respond to the “reaction” level of Kirkpatrick's
Kontoghiorghes [25] should be employed. four-level theory of training evaluation, [23,24] the
Therefore, the objective of this paper is to first level before learning, behaviors, and results.
answer and explain which learning environment With similar idea introduced by the classical diffusion
factors and individual factors affect students’ theory, users evaluate the advantages of a new
e-learning effectiveness. To answer the technology, bringing positive attitude towards the
above-mentioned issue, we will follow the research system, before they continuously using the new
procedure to answer the following questions: How to technology.
define students’ e-learning effectiveness? What are In the classical diffusion theory, moreover,
the main factors that affect students’ e-learning “use” is a main variable for effectiveness assessment
effectiveness? Of these factors, which are the factors of the introduction of new technologies and systems.
at the college/university environment level? Which The main difference between e-learning and
are the factors at the individual level? How do these traditional learning models lies in application of
factors affect students’ e-learning? digital technologies. Researchers of students’
Altogether, this study aims to: (1) apply the e-learning adopt the view of technology use and
socio-technical systems theory to review and consider the uniqueness of technology use in
integrate theories about student e-learning from a students’ e-learning, suggesting that “use of
macro view and construct the research framework of e-learning” should be included in the e-learning
this study; (2) construct a multi-level analysis model framework. [10,11]. Students get a chance to learn
and propose hypotheses, through literature review; (3) from e-learning system after the system was used by
verify the theoretic model and relationship between students. Therefore, after the reaction level – usability,
research variables through multi-source and “use of e-learning” should be the formative key factor
multi-stage questionnaire survey and multi-level of the summative variable of e-learning effectiveness
analysis of the data; and (4) propose conclusions - e-learning performance, referring to knowledge and
based on empirical evidence and explain factors skills getting from learning through e-learning for the
affecting students’ e-learning effectiveness and how blended course.
to design and manage e-learning.
2.2 Socio-technical Systems and a Multi-level
2. LITERATURE REVIEW Theoretic Framework
Falconer [12] points out that organizational
2.1 Students’ e-Learning and Evaluation of Its learning involves multiple levels of an organization.
Effectiveness In an integrative review of literature associated with
In terms of American Society of Training and e-learning effectiveness, researchers have proposed
Education [1], e-learning refers to the use of that a systematic and multi-level consideration of
electronic devices for learning, including the delivery students’ e-learning effectiveness is necessary [7,33]
of content via electronic media such as internet, audio and the socio-technical systems approach adopted by
or video, satellite broadcast, interactive TV, CD-ROM, Kontoghiorghes [25] should be employed. In this
and so on. And the application of e-learning includes paper, we apply the socio-technical systems approach
online learning, virtual classrooms and digital to review and integrate factors affecting students’
collaboration, etc. e-learning effectiveness from a macro view and
Based on Newstrom’s [31] procedural view as construct a multi-level analysis model. The
well as the summative and formative indexes of socio-technical system consists of two interdependent
training effectiveness and the classical diffusion subsystems, namely the social system and the
theory, we propose a three-level framework of technical system. The social system encompasses
students’ e-learning effectiveness, as shown in Figure organizational culture, interpersonal relations, values,
1. beliefs, motivations, interactive patterns, learning,
and adaptability to changes. The technical system
Re a ct io n Us e Le a rn in g encompasses mechanical facilities, technical methods,
and professional knowledge. [30,35,47]
Us e o f E- Le a rn in g
Us a b ilit y
E- le a rn in g Pe rfo rm a n ce However, in a multi-level framework, how do
the technical system and the social system influence
Figure 1: The three-level framework of students’
students’ e-learning effectiveness? Salmon [42]
e-learning effectiveness
discusses e-learning in a book titled e-Moderating,
314 International Journal of Electronic Business Management, Vol. 8, No. 4 (2010)

emphasizing that digital technology is an important 3. CONCEPTUAL FRAMEWORK,

“moderator” in learning activities, as it is a condition
and an environment that can assist learners in HYPOTHESES, AND RESEARCH
learning activities. Besides, individual applications of DESIGN
technology are also influenced by effects of social
relations that exist between people, such as the According to literature review, students’
reward system and power structure [48]. Chu and e-learning effectiveness can be divided into three
Chu [7] point out that the extant empirical research levels, which are respectively dominated by three
on students’ e-learning effectiveness does not fully variables, including usability, use of e-learning and
support a positive relationship between e-learning and e-learning performance. We identify a set of key
learning effectiveness. They argue that there may be individual and learning-environment factors, which
moderators between e-learning and learning impacts these three dependent variables of e-learning
effectiveness, so further exploration of the effectiveness mainly based on theoretical streams. As
moderators is necessary. Therefore, we employ the shown in Figure 2, ease of use is referred to
socio-technical systems view to construct a ‘technology acceptance model (TAM)’, while media
multi-level analysis framework, with a focus on the richness is referred to the ‘media richness theory’.
effects of each subsystem at the individual They both follow the stream of the classical diffusion
level–students’ e-learning effectiveness. theory. Learning attitude, extrinsic motivation and
learning climate are referred to ‘social learning
theory’, and motivation theories. We also have
verified the variables and relationships after
interviews of students from different blended courses
in colleges. Hypotheses are inferred and proposed in
the following paragraphs.

Te ch n ica l Co n d it io n So cia l Su p p o r t Co n d it io n

Ea s e o f
Us e H1
Ext rin s ic Le a rn in g
H2 Us a b ilit y
M o t iva t io n Clim a t e
M e d ia
Ric h n e s s

Le ve l 2
H4
Le a rn in g En viro n m e n t

Le ve l 1 H6 H7 H11
H5
St u d e n t H10
H9
Le a rn in g Us e o f E- le a rn in g
At t it u d e H3 E- le a rn in g H8 Pe r fo rm a n ce

Figure 2: Conceptual framework

3.1 Usability of e-Learning Studies of the Internet also point out that the rich
In terms of technology acceptance model information media embedded in websites can enhance
(TAM) [10], perceived ease of use is the antecedent satisfaction of website users (e.g., [38]).
of perceived usefulness. This idea has been validated, Lee [26] indicated that media richness is
especially in empirical study of e-learning [39,49]. important factor of e-learning. Hackman and Walker
And according to the media richness theory [9], [14] also indicated that media richness contributes to
media richness refers to using multiple and most increasing e-learning satisfaction after the empirical
appropriate digital media to present learning content. study of the use of television for teaching.
Therefore, we propose: Researchers, such as Chen et al. [4], Liu et al. [29]
H1: Perceived ease of use positively influences and Sun & Cheng [46], all indicates that media
usability of e-learning system. richness will affect learner’s satisfaction and the
Many studies of e-learning have revealed that assessment of e-learning system usability. Therefore,
application of a plurality of media [21] and we propose:
interactive media [16] can draw learners’ attention H2: Media richness positively influences usability
and effectively enhance their learning effectiveness. of e-learning system.
W. C. Wu and L. Y. Hwang: The Effectiveness of e-Learning for Blended Courses in Colleges 315

3.2 Use of e-Learning students’ use of e-learning.

In a learning environment, “motivation” refers
to the incentive that propels students to be devoted to Both intrinsic and extrinsic motivation of
learning activities and intrinsic motivation is the learning is very important in students’ engagement in
critical success factor of e-learning. Ryan et al. [40] the learning experiences. Extrinsic motivation can be
have developed the self-determination theory and defined as external factors that stimulate learners and
proposed the subsequent research that motivation is a these external factors can be behaviors of teachers,
continuum of self-determination levels which reflect learning topics, learning-teaching strategies,
the perceived locus of causality. Motivations in the teaching-learning process, interaction between the
highest level are intrinsic motivations (enjoy doing it) students and teachers and so on [44]. Extrinsic
and extrinsic motivations including autonomous motivation is one way that encourages students to
motivation (worth doing) and controlled motivation commit themselves to instructional goals to increase
(supposed to do it; must do it). Learning attitude their achievement such as earning a grade or degree
stands for the behavior of one who likes the course [43, 45]. Therefore for blended courses when teachers
he/she takes, and enjoys participating in the activities provide encouragement, praise, reward, and higher
of this course. Therefore, for the blended course, scores in grades for use e-learning, these extrinsic
good learning attitude may also turn into an intrinsic motivations will increase students’ use of e-learning.
motivation, the enthusiasm, for using e-learning for H6: Extrinsic motivation positively influences
participating in the activities of the course. In the students’ use of e-learning.
study of e-learning, Saade et al. [41] indicated when
students’ intrinsic motivation is stronger, their Learning climate is defined as the learning
willingness to use e-learning will be higher. Therefore, atmosphere among all students of the same course. A
we propose: positive learning climate encourages and stimulates
H3: Learning attitude positively influences students’ the exchange of ideas, opinion, information, and
use of e-learning. knowledge [37]. According to social learning theory,
learners will imitate others’ learning behaviors via
Besides, many empirical studies conducted on observing their learning activities. The more students
the basis of the classical diffusion theory have take part in e-learning activities for their blended
pointed out that users’ positive attitude to a technical course, will the more others in the same course be
system can help to increase their intention to use the encouraged to join them. Prior research [36] shows
system. Ong et al. [32] and Chiu et al. [5] also that social interaction has a direct effect on the usage
discovered through a research of students’ e-learning of an e-learning system, and the study of Chou and
that increase of user satisfaction will positively Liu [6] has indicated that learning climate is an
influence students’ intention to continue using the important antecedent of using an e-learning system.
e-learning system. In this paper, the construct of Therefore, we propose:
usability of e-learning system encompasses students’ H7: Learning climate positively influences students’
perceived satisfaction, effectiveness, and efficiency, use of e-learning.
so we propose:
H4: Usability of e-learning system positively 3.3 e-Learning Performance
influences students’ use of e-learning. The main difference between e-learning and
conventional learning models lies in application
Information systems can help increase learning (usage) of digital technologies. Researchers of
efficiency [15,19] and technical systems play the role e-learning adopt the view of technology use and
of an “e-moderator” [42]. A useful e-learning system consider the uniqueness of technology use in
can provide a better system environment for learning, e-learning, suggesting that “use” should be
allowing students of better learning attitude, the encompassed in the e-learning framework [10,11].
intrinsic motivation, to learn in a good environment Transfer of teaching via e-learning system requires
and meet the requirements for their blended course. students to use e-learning in a self-directed manner;
Learning content can more effectively enhance otherwise, the learning effectiveness can’t be
students’ intention to use e-learning when transferred achieved. The more students using e-learning, the
by a better e-learning system. Hence, satisfaction of more they may get from this alternative of learning
an e-learning system, better usability, can reinforce activities. Therefore, we propose:
the “motivation to use” induced by better learning H8: Students’ use of e-learning positively influences
attitude, propelling students to have more intention to e-learning performance.
continue using e-learning. Based on the above
discussion, the following hypothesis is proposed. Usability encompasses users’ perceived
H5: Usability of e-learning system reinforces the satisfaction, effectiveness, and efficiency towards the
positive influence of learning attitude on e-learning system, and in other words, the reaction
316 International Journal of Electronic Business Management, Vol. 8, No. 4 (2010)

level of Kirkpatrick's four-level framework of students’ e-learning performance was measured by

training evaluation. [23,24] The research of training self-evaluation (questions for example, “I have made
transfer indicates that trainee’s “reaction” to training, a breakthrough in understanding the learning subjects
i.e. satisfaction, affects training outcomes (e.g., [2]). through e-learning.”), which may distinguish
Studies based on the theory of reasoned action (TRA) e-learning performance from overall learning
and TAM also have a similar argument that user’s outcome they get from their blended course. Principal
reaction to technology use affects user performance component analysis of exploratory factor analysis
[10]. The empirical evidence in Johnson et al. [20] (EFA) was conducted to verify the construct validity
suggests that usability has positive effects on of the questionnaire. Results shows the factors were
e-learning outcomes, including e-learning as the constructs expected, and all of good factor
performance. Therefore, we propose: loadings. Besides, according to the factor analysis
H9: Usability of e-learning system positively results, all the constructs were tested for reliability,
influences students’ e-learning performance. results showing all greater than 0.8 (0.834 - 0.932).

Students’ individual activities may be subject to

the influence of the learning environment. Chu and
4. DATA ANALYSIS AND
Chu [7] indicated that previous studies haven’t fully RESULTS
supported the positive relationship between
e-learning and learning effectiveness. Besides, there 4.1 Aggregation Statistics
may be important moderators between them, so This research was designed to conduct a
further exploration is required. We argue that the multi-level analysis. Because many variables were
social supports from learning environment, including induced from data of many individuals, theoretically,
“extrinsic motivation” and “learning climate” are there should be a certain degree of consistency or
influential. consensus between individuals. Therefore,
In addition to students’ use of e-learning, consistency between individuals in the perception of
connecting goal settings and reward plans to a specific phenomenon is a key criterion that
e-learning performance can increase the determines whether a group variable can be generated.
instrumentality of e-learning and thus enhance the Besides, variance of the variable should be tested by
motivation for e-learners. Besides, extrinsic explaining the group variable and measuring the
motivations provide learning utility, such as praises, mean reliability at the group level. We conducted the
rewards, identification, encouragement, and higher test by measuring rwg (within group agreement),
scores in grades, and can elevate the value of intra-class correlation coefficient (ICC(I)), and
e-learning behaviors. Researchers of the cognitive reliability at the group level (ICC(II)). The results
evaluation theory argue that extrinsic motivation and show that all the rwg values (0.61 – 0.92) are greater
intrinsic motivation have interaction effects on than the 0.6 threshold suggested by James [18].
behaviors [3]. Thus, we propose following
hypotheses: Table 1: The effects on usability
H10: Extrinsic motivation reinforces the positive Dependent: Usability
Model Model
influence of students’ use of e-learning on Independent
A1 A2
e-learning performance. Ease of Use 0.886 *** 0.520 ***
H11: Learning climate reinforces the positive Media Richness 0.469 ***
influence of students’ use of e-learning on R2 0.786 0.872
ΔR2 0.086 ***
e-learning performance. F 110.05 *** 98.91 ***
* p<0.05；** p<0.01；*** p<0.001
3.4 Participants, Procedures, and Measures
To verify the proposed hypotheses and the With Hierarchical Linear Modeling (HLM)
overall model, we invited 32 classes of analysis, estimated on the basis of the null model
undergraduates in National Taipei University in (Model B1 in Table 2), ICCI(I) of “use of e-learning”
Taiwan. The questionnaire was collected in one is 0.13; and estimated on the basis of the null model
month at the end of the second semester in 2009. (Model C1 in Table 3), ICC(I) of “e-learning
After 43 responses with missing or incomplete performance” is 0.14. Both ICC(I) values are greater
answers were excluded, a total of 1227 valid students’ than the 0.059 standard proposed by Cohen [8].
responses from 32 classes were obtained. Besides, both of the ICC(II) values (0.802 for use of
All measures (with 3-4 items) of constructs e-learning; 0.812 for e-learning performance) are
were modified from measures of former empirical greater than the 0.6 standard proposed by Glick [13].
studies, developed according to previous empirical
studies of good reliabilities (e.g. [28,34]) and
modified on the basis of related theories. In this study,
W. C. Wu and L. Y. Hwang: The Effectiveness of e-Learning for Blended Courses in Colleges 317

4.2. Hypotheses Testing Results as well as extrinsic motivation (γ =0.308, p<0.1)

＾

Results for usability. For small sample size of from level 2. Students’ use of e-learning for blended
the group level in this study, we apply the courses is positively influenced by students’ learning
significance level of p-value = 0.1. H1 and H2 predict attitude, the usability of e-learning system, and the
that perceived ease of use and media richness will be extrinsic motivation in class. Besides, with better
associated with the usability of e-learning system. usability of e-learning system, students of better
The result of Model A2 in Table 1 presents ease of learning attitude will more likely use e-learning for
use (β=0.520, p<0.001) and media richness (β=0.469, blended course. However, learning climate is not
p<0.001) are significantly related to usability of significantly related to use of e-learning. Therefore,
e-learning, providing support to H1 and H2. the results support H3 to H6, but do not support H7.
Therefore, both perceived ease of use and media Results for e-learning performance. With
richness influences the usability of students’ HLM models, we followed a similar procedure in
e-learning for blended courses in colleges. testing H8 to H11. H8 and H9 predict that students’
Results for use of e-learning. To test H3 to use of e-learning (level 1) and the usability of
H7, we estimated HLM models in which students’ e-learning system (level 2) will be associated with
learning attitude was the level 1 (student level) students’ e-learning performance, while H10 and H11
predictor (Model B3 in Table 2), after computer self predict that extrinsic motivation and learning climate
efficacy been controlled, Model B2, and then in class will reinforce the influence from students’ use
regressed the intercept coefficients obtained from of e-learning to their e-learning performance. As
level 1 on the measures of learning-environment- Model C5 shown in Table 3, the results reveal that
level usability of e-learning, learning climate, and
both students’ use of e-learning (γ =0.380, p<0.001)
＾

extrinsic motivation in class at level 2 (Model B4).

and the usability of e-learning system (γ =0.898,
＾
Meanwhile, we estimated the cross-level reinforcing
effect from usability of e-learning system to the p<0.001) are positively related to students’ e-learning
relationship between students’ learning attitude and performance. Moreover, the cross-level interactions
their use of e-learning (Model B5). As Model B5 of extrinsic motivation ( γ =0.114, p<0.05) and
＾

reported in Table 2, students’ learning attitude and

learning climate (γ =0.170, p<0.1) in class are both
＾

their use of e-learning has a significantly positive

relationship ( γ =0.364, p<0.001), which is also
＾ significant, showing that both of these two learning-
environment variables reinforce the relationship
moderated (reinforced) by the usability of e-learning between students’ use and their final performance of
system (cross-level effect, γ =0.356, p<0.01).
＾
e-learning. Hence, H8 to H11 are supported.
Usability also has a directly positive relationship
(γ =0.522, p<0.001) with students’ use of e-learning,
＾

Table 2: The effects on use of e-learning

Dependent: Use of E-learning
Model B5
Model B1 Model B2 Model B3 Model B4
Independent Interaction
Null Model Control Students Environment
Effects
Level 1
(Constent) 3.116 *** 3.107 *** 3.099 *** 3.117 *** 3.103 ***
Computer Self Efficacy 0.246 *** 0.143 ** 0.137 ** 0.135 **
Learning attitude 0.347 *** 0.333 *** 0.364 ***
Level 2
Usability 0.583 *** 0.522 ***
Learning Climate 0.165 0.168
Extrinsic Motivation 0.293 † 0.308 †
Cross-Level
Learning Attitude ×
0.356 **
Usability
Between-Group
0.118 *** 0.096 *** 0.078 *** 0.037 *** 0.035 ***
Residual Variance
Within-Group
0.746 0.714 0.673 0.672 0.667
Residual Variance
R2within-group 0.100
R2between-group 0.686 0.703
3127.3
Model Deviance 3182.96 3056.63 3039.44 3034.81
7
† P<0.1; * P<0.05; ** P<0.01; *** P<0.001
318 International Journal of Electronic Business Management, Vol. 8, No. 4 (2010)

Table 3: The effects on e-learning performance

Dependent: E-learning Performance
Model C5
Model C1 Model C2 Model C3 Model C4
Independent Interaction
Null Model Control Students Environment
Effects
Level 1
(Constent) 3.471 *** 3.464 *** 3.440 *** 2.013 *** 3.489 ***
Computer Self Efficacy 0.183 *** 0.091 ** 0.089 ** 0.085 ***
Use of E-learning 0.377 *** 0.359 *** 0.380 ***
Level 2
Usability 0.632 *** 0.898 ***
Cross-Level
Use of E-learning ×
0.114 *
Learning Climate
Use of E-learning ×
0.170 †
Extrinsic Motivation
Between-Group
0.075 *** 0.064 *** 0.037 *** 0.006 * 0.002 *
Residual Variance
Within-Group
0.455 0.436 0.338 0.338 0.335
Residual Variance
R2within-group 0.257
R2between-group 0.920 0.973
2213.9
Model Deviance 2577.76 2528.09 2184.57 2160.10
1
† P<0.1; * P<0.05; ** P<0.01; *** P<0.001

4.3. Discussion develop and improve students’ e-learning

Among all hypotheses, H7 is not supported by performance.
the result of HLM analysis. That means learning Usability of an e-learning system is an
climate, referring to learning condition of students in important determination of students’ use of
a class as a whole, doesn’t seem directly to impact e-learning and the performance. A well-designed
students’ use of e-learning. For particularity of e-learning system can not only enhance students’
e-learning, “autonomy and independence” exist in intention to use the system but also help them obtain
most circumstances when students use e-learning, learning effectiveness. This finding is consistent with
while far away from the class, and also their conclusions of many studies on technology use and
classmates. This may explain why the influence of e-learning. However, through a multi-level research
learning climate is not significant to students’ use of framework, we explore the moderating role of
e-learning. technology at the learning environment level. The
finding indicates that students’ use of e-learning
5. CONCLUSION AND induced by the learning attitude can amplify the
effects of a good e-learning system. Our empirical
SUGGESTIONS evidence also supports Salmon’s [42] argument about
the moderator of e-learning in e-Moderating.
Responding to social-technical system theory, Ease of use and media richness of e-learning
factors of technical system and social support are key indicators in students’ evaluation for the
system simultaneously or alternately affect usability of an e-learning system. “What kind of
effectiveness of students’ e-learning. This stands for e-learning system is a satisfactory learning system?”
the viewpoints of existing social-technical studies is a focal issue in many existing studies. Through
(e.g. [25,33]). Here in this study, we have illustrated literature review, we induced the development
usability of e-learning and its antecedents, ease of use process of this research domain. In early years, the
and media richness, are essential factors of the research of e-learning technologies was focused on
technical system; while extrinsic motivation and system quality and service quality of the supporting
learning climate are important factors of the social systems. Later, researchers began to shift their focus
system. These factors directly impact or moderate the onto information quality, i.e. quality of e-learning
influence to students’ use of e-learning and their materials. In recent years, compared with the system
e-learning performance. Therefore, managers of aspect, researchers paid more attention to the content
e-learning in universities need to take both technical aspect, i.e. sufficiency, richness, and presentation of
factors and social factors into account while trying to the learning content. Therefore, many recent
W. C. Wu and L. Y. Hwang: The Effectiveness of e-Learning for Blended Courses in Colleges 319

researchers have employed the media richness theory the present study we identified a set of key individual
proposed by Daft and Lengel [9] to propose the and contextual correlates of students’ e-learning
importance of content richness and use of multimedia effectiveness for blended courses after interviews of
in e-learning. Findings in our study also support this college students. But, there are likely to be other
viewpoint. factors that have impacts. For example, value,
The better learning attitude, the students are adaptability to changes related to social system, and
more earnest in use of e-learning, especially with facilities and professional knowledge related to
an e-learning system of better usability. Taking technical system are additional factors to be
e-learning as an alternative to get knowledge, considered in the future. However, compared with the
intrinsic motivation of students devoted to learning existing research, this paper uses a more
activities is the critical success factor of e-learning. comprehensive system view to construct the
This finding also stands for the study of Saade et al. theoretical model and empirically verify it. The
[41] which indicated when students’ intrinsic results can be a reference for future researchers and
motivation is stronger, their willingness to use managers of e-learning in colleges.
e-learning will be higher. Meanwhile, to meet the
demand for higher learning efficiency, a useful REFERENCES
e-learning system can provide a better system
environment for learning, allowing students with 1. American Society of Training and
intrinsic motivations of learning to learn in a good Development (ASTD), 2001, A vision of
environment. e-learning for America’s workforce: Report of
Extrinsic motivation encourages students to the Commission on Technology and Adult
use e-learning, and provides incentive to learn Learning, ED455432, Alexandria, VA: ASTD.
better on e-learning as well. Extrinsic motivations, 2. Baldwin, T. T. and Ford, J. K., 1988, “Transfer
including encouragement, praise, reward, and higher of training: A review and direction for future
scores in grades, serve as an effective “situational” research,” Personnel Psychology, Vol. 41, pp.
factor (strong instrumentality of the expectancy 63-105.
theory) that not only modifies students’ behavior to 3. De Charms, R., 1968, Personal Causation: The
use e-learning, but also enhances students’ learning Internal Affective Determinants of Behavior,
motivation to learn well. New York: Academic Press.
Learning climate does not directly impact 4. Chen, C. C., Wu, J. and Yang, S. C., 2006,
students’ use of e-learning, but provides “The efficacy of online cooperative learning
advantageous environment for students to systems: The perspective of task-technology
improve e-learning performance. Many studies of fit,” Campus-Wide Information Systems, Vol.
23, No. 3, pp. 112-127.
learning theories have pointed out that “learning
5. Chiu, C., Hsu, M., Sun, S., Lin, T. and Sun, P.,
climate” in a class affects students’ behaviors
2005, “Usability, quality, value and e-learning
inconspicuously. From the learning perspective, the
continuance decisions,” Computers &
social learning theory also provides a foundation for Education, Vol. 45, pp. 399-416.
this argument. Compared to other direct factors in the 6. Chou, S. W. and Liu, C. H., 2005, “Learning
current study, such as usability, extrinsic motivation, effectiveness in a web-based virtual learning
and learning attitude, learning climate is not environment: A learner control perspective,”
significantly directly influence students’ behavior of Journal of Computer Assisted Learning, Vol.
using e-learning, but it still builds social motivation 21, pp. 65-76.
for those of using e-learning to learn more. 7. Chu, R. J. and Chu, A. Z., 2010, “Multi-level
A “useful” e-learning system should be defined analysis of peer support, Internet self-efficacy
by users. Managers of e-learning in universities may and e-learning outcomes–The contextual effects
collect responds from both students and lecturers of collectivism and group potency,” Computers
regularly, to modify the system for meeting the need & Education, Vol. 55, pp. 145-154.
of various classes. Moreover, to improve 8. Cohen, J., 1988, Statistical Power Analysis for
effectiveness of e-learning in universities is not only a the Behavioral Sciences, (2nd ed.), Hillsdale,
technical issue. The lecturers, who may provide NJ: Eribaum.
extrinsic motivation and developing a learning 9. Daft, R. L. and Lengel, R. H., 1984,
climate in their classes, play an indispensable role for “Information richness - a new approach to
students to use e-learning constantly and earnestly. managerial behavior and organization design,”
Limitations and Future Research. Due to the Research in Organizational Behavior, Vol. 6,
constraint of research needs, a non-random sampling pp. 191-233.
method and self-report survey were adopted. As a 10. Davis, F. D., 1986, A Technology Acceptance
Model for Empirically Testing New End-User
result, there was limitation on generalization of the
Information System: Theory and Results,
theories. Moreover, drawing on existing research, for
320 International Journal of Electronic Business Management, Vol. 8, No. 4 (2010)

Doctoral dissertation, Sloan School of Resource Development, (3nd ed.), New York,
Management, MIT, Cambridge, MA. NY: McGraw-Hill.
11. DeLone, W. H. and McLean, E. R., 2003, “The 25. Kontoghiorghes, C., 2004, “Reconceptualizing
DeLone and McLean model of information the learning transfer conceptual framework:
systems success: A ten-year update,” Journal of Empirical validation of a new systemic model,”
Management Information Systems, Vol. 19, No. International Journal of Training and
4, pp. 9-30. Development, Vol. 8, No. 3, pp. 210-221.
12. Falconer, L., 2006, “Organizational learning, 26. Lee, Y., 2006, “An empirical investigation into
tacit information, and e-learning: A review,” factors influencing the adoption of an
The Learning Organization, Vol. 13, No. 2, pp. e-learning system,” Online Information Review,
140-151. Vol. 30, No. 5, pp. 517-541.
13. Glick, W. H., 1985, “Conceptualizing and 27. Liaw, S. S., Huang, H. M. and Chen, G. D.,
measuring organizational and psychological 2007, “Surveying instructor and learner
climate: Pitfalls in multilevel research,” attitudes toward e-learning,” Computers
Academy of Management Review, Vol. 10, pp. Education, Vol. 49, No. 4, pp. 1066-1080.
601-616. 28. Lim, H., Lee, S. and Nam, K., 2007,
14. Hackman, M. Z. and Walker, K., 1990, “Validating e-learning factors affecting training
“Instructional communication in the televised effectiveness,” International Journal of
classroom,” Communication Education, Vol. 39, Information Management, Vol. 27, pp. 22-35.
pp. 145-156. 29. Liu, S. H., Liao, H. L. and Pratt, J. A., 2009,
15. Hauser, J. R. and Clausing, D., 1988, “The “Impact of media richness and flow on
house of quality,” Harvard Business Review, e-learning technology acceptance,” Computers
Vol. 67, pp. 63-73. & Education, Vol. 52, pp. 599-607.
16. Hunt, B., Burvall, P. and Ivergard, T., 2004, 30. Molleman, E. and Broekhuis, M., 2001,
“Interactive media for learning (IML): Assuring “Sociotechnical systems: towards and
usability in terms of a learning context,” organizational learning approach,” Journal of
Education and Training, Vol. 46, No. 6/7, pp. Engineering and Technology Management, Vol.
361-369. 18, pp. 271-294.
17. Islas, E., Perez, M., Rodriguez, G., Paredes, I., 31. Newstrom, J. W., 1978, “Catch-22: The
Avila, I. and Mendoza, M., 2007,“E-learning problems of incomplete evaluation of training,”
tools evaluation and roadmap development for Training and Development Journal, Vol. 32, pp.
an electrical utility,＂ Journal of Theoretical 22-24.
and Applied Electronic Commerce Research, 32. Ong, C., Lai, J. and Wang, Y., 2004, “Factors
Vol. 2, No. 1, pp. 63-75. affecting engineers’ acceptance of
18. James, L. R., 1982, “Aggregation bias in asynchronous e-learning systems in high-tech
estimates of perceptual agreement,” Journal of companies,” Information & Management, Vol.
Applied Psychology, Vol. 67, pp. 219-229. 41, No. 6, pp. 795-804.
19. Jessup, L. M. and Valacich, J. S., 1993, Group 33. Ozkan, S. and Koseler, R., 2009,
Support Systems–New Perspectives, New York: “Multi-dimensional students’ evaluation of
Macmillan. e-learning systems in the higher education
20. Johnson, R. D., Hornik, S. and Salas, E., 2008, context: An empirical investigation,”
“An empirical examination of factors Computers & Education, Vol. 53, pp.
contributing to the creation of successful 1285-1296.
e-learning environments,” International 34. Park, J. and Wentling, T., 2007, “Factors
Journal of Human-Computer Studies, Vol. 66, associated with transfer of training in
pp. 356-369. workplace e-learning,” Journal of Workplace
21. Keller, J. M., 2008, “First principles of Learning, Vol. 19, No. 5, pp. 311-329.
motivation to learn and e3-learning,” Distance 35. Pasmore, W. A., 1988, Designing Effective
Education, Vol. 29, No. 2, pp. 175-185. Organizations: The Socio-technical Systems
22. Kim, K. K., 1989, “User satisfaction: A Perspective, New York, NY: Wiley.
synthesis of three different perspectives,” 36. Pituch, K. A. and Lee, Y., 2006, “The influence
Journal of Information Systems, Vol. 4, No. 1, of system characteristics on e-learning use,”
pp. 1-12. Computers & Education, Vol. 47, pp. 222–244.
23. Kirkpatrick, D. L., 1967, “Evaluation of 37. Prieto, I. M. and Revilla, E., 2006, “Formal and
training,” In: R.L. Craig & L.R. Bittel, Training informal facilitators of learning capability: The
and Development Handbook, New York, NY: moderating effect of learning climate,”
McGraw-Hill. WP06-09, Universidad de Valladolid, Spain.
24. Kirkpatrick, D. L., 1987, “Evaluation of 38. Qina, J., Zhoub, Y., Reidc, E., Laid, G. and
training,” In: R.L. Craig, Training and Chenc, H., 2007, “Analyzing terror campaigns
Development Handbook: A Guide to Human on the internet: Technical sophistication,
W. C. Wu and L. Y. Hwang: The Effectiveness of e-Learning for Blended Courses in Colleges 321

content richness, and Web interactivity,” ABOUT THE AUTHORS

International Journal of Human-Computer
Studies, Vol. 65, pp. 71-84. Wenchieh Wu, Doctoral Lecturer of Department of
39. Roca, J. C., Chiu, C. and Martinez, F. J., 2006, Business Administration, National Taipei University,
“Understanding e-learning continuance Taiwan. He received a Ph. D. in Business
intention: An extension of the Technology Administration from National Taipei University. His
Acceptance Model,” International Journal of current research interests include organizational
Human-Computer Studie, Vol. 64, pp. 683-696. learning, knowledge management, leadership, and
40. Ryan, R. M., Connell, J. P. and Deci, E. L., customer behavior. Dr. Wu has published papers in
1985, “Amotivational analysis of refereed journals such as: NTU Management Review,
self-determination and self-regulation in Marketing Review, and Journal of Tungnan
education,” In: C. Ames and R.E. Ames, University.
Research on Motivation in Education: The
Classroom Milieu, New York, NY: Academic. Lan-Yin Hwang, MBA student of Department of
41. Saade, R. G., He, X. and Kira, D., 2007, Business Administration, National Chengchi
“Exploring dimensions to online learning,” University, Taiwan.
Computers in Human Behavior, Vol. 23, pp.
1721-1739. (Received May 2010, revised August 2010, accepted
42. Salmon, G., 2004, E-Moderating -The Key to September 2010)
Teaching and Learning Online, (2nd ed),
London, UK: Routledge Falmer.
43. Selvi, K., 2006, “Learning and creativity,” In:
A. T. Tymienecka, Analecta husserliana: The
yearbook of phenomenological research,
Dordecth: Springer.
44. Selvi, K., 2010, “Motivating factors in online
courses,” Procedia Social and Behavioral
Sciences, Vol. 2, pp. 819-824.
45. Styer, A. J., 2007, “A grounded meta-analysis
of adult learner motivation in online learning
from the perspective of the learner,” Capella
University, Unpublished Dissertation Thesis.
46. Sun, P. C. and Cheng, H. K., 2007, “The design
of instructional multimedia in e-Learning: A
Media Richness Theory-based approach,”
Computers & Education, Vol. 49, pp. 662-676.
47. Trist, E., 1981, The Evaluation of
Sociotechnical Systems, Toronto: Quality of
Working Life Center.
48. Truffer, B., 2008, “Society, technology, and
region: Contributions from the social study of
technology to economic geography,”
Environment and Planning A, Vol. 40, pp.
966-985.
49. Wang, Y., Wang, H. and Shee, D., 2007,
“Measuring e-learning systems success in an
organizational context: Scale development and
validation,” Computers in Human Behavior,
Vol. 23, pp. 1792-1808.
 International Journal of Electronic Business Management, Vol. 8, No. 4 (2010) 322

大學混成課程之數位學習成效：
跨層次實證研究
吳文傑 1*、黃蘭茵 2
1
國立臺北大學企業管理學系
台北縣三峽鎮大學路 151 號
2
國立政治大學企業管理學系
台北市文山區指南路二段 64 號

摘要
當數位學習成為大學傳統教學外多元的強大工具，教與學就不再侷限在傳統教室內
了。隨著科技的發展並以網際網路為傳遞平台，大學校院願意投資資源在發展線上課
程，同時，結合持續性支持系統之線上內容與傳統面對面課堂內容的混成課程，已快
速成長為教與學的新選擇。然而，數位學習的發展與管理仍具挑戰，是機會但也有阻
礙，特別是針對透過數位學習進行混成課程之學生的學習成效。本研究以宏觀角度，
應用社會技術系統理論，探討並整合學生數位學習的相關理論。為補充相關研究之不
足，首先透過文獻探討瞭解影響數位學習成效之前因，以建立分析模式；接著，透過
已使用數位學習之學生問卷調查並進行跨層次分析，驗證影響因素間關係的假說及研
究模式。結果顯示數位學習成效（數位學習易用性、使用，及數位學習績效）交替或
同時受到來自學習環境層次與個人層次中，技術系統與社會系統的直接或干擾因素的
影響。比較現有研究，本文運用廣泛的系統觀點建構理論模式並以實證檢驗，研究結
果可供未來研究者及大學數位學習管理者參考。

關鍵詞：數位學習、社會技術系統、跨層次分析
（*聯絡人：sebwooo@gmail.com）