Shih, K.-P., Chen, H.-C., Chang, C.-Y., & Kao, T.-C. (2010).

The Development and Implementation of Scaffolding-Based Self-

Regulated Learning System for e/m-Learning. Educational Technology & Society, 13 (1), 80–93.

The Development and Implementation of Scaffolding-Based Self-Regulated

Learning System for e/m-Learning
Kuei-Ping Shih1, Hung-Chang Chen2, Chih-Yung Chang1* and Tai-Chien Kao3
1
Department of Computer Science and Information Engineering, Tamkang University, Tamsui, Taipei County,
Taiwan // Tel: +886-2-26215656 Ext. 2748 // [email protected] // *[email protected]
2
Department of Information Technology, Ching Kuo Institute of Management and Health, Keelung, Taiwan // Tel:
+886-2-24372093 Ext. 273 // [email protected]
3
Institute of Education, National Dong Hwa University, Hualien, Taiwan // Tel: +886-3-8635571 Ext. 5571 //
[email protected]

ABSTRACT
This paper proposes a self-regulated learning (SRL) system with scaffolding support in order to develop
independent learning skills among students. The SRL system uses self-regulated learning and scaffolding
theories to appeal to both instructors and learners. On the part of the instructors, a Content Accessibility
Subsystem is provided to easily organize learning materials and to dynamically provide different levels of support
for their learners. As for the learners, many subsystems are proposed that provide a conducive mobile learning
environment for them. With the application of the scaffolding theory, the system can easily adjust to provide help
to the learners, facilitating SRL processes anytime and anywhere, and establishing the learners’ SRL patterns
gradually. The learners in the experiment deemed that that the proposed system could provide them self-
regulatory attributes. The experiment results show that the average SRL score of learners increases, though the
improvement is not significant. However, the result also showed that the SR skills of students in the group of
Low SR significantly improved.

Keywords
Self-Regulated Learning, Self-Regulatory Learning Cycle, Scaffolding, Mobile Leaning, E-Learning, CAL systems

Introduction
The main goal of education is to develop the character of students and foster in them a spontaneous desire to learn.
To achieve this aim, self-regulated learning (SRL) is essential. However, while modern technologies have made
learning possible at any time and place, there still is the challenge to provide a conducive environment so that
learners can easily schedule their study plans and avail of learning materials outdoors. To address this, this paper
proposes an SRL system with wireless technologies.

Lately, people from both academic and government sectors have keenly promoted SRL because they recognize the
need to help learners take charge of their own education. However, SRL is not an easy task. Four factors are essential
in carrying out SRL: learning schedules, materials, scenarios, and quality (Zimmerman, Bonner, & Kovach, 1996).
Along with these factors, other difficulties in performing SRL may be pointed out:
 Learning schedules and materials: A suitable learning schedule makes a person’s own learning methodical.
However, if there is insufficient experience in the design of a learning schedule, a person may end up having
poor SRL performance. In addition, because the study materials are limited and varied, a learner may find it
difficult to organize the materials he needs.
 Learning scenarios and quality: The rapid development of modern technologies, such as broadband and wireless
communication engineering, makes learning materials easily available. However, because there is no tailor-
made learning environment for outdoor scenarios, learners may just give up learning due to the difficulty or the
complexity of accessing the learning materials. Moreover, because of many distractions, learners may be unable
to focus well.

Therefore, an SRL system that adopts the concept of a self-regulatory learning cycle (Zimmerman et al., 1996) is
proposed. Because having an ambitious and unrealistic aim may disappoint learners during the process of learning,
the proposed system firstly helps students set a reasonable goal in initiating their motive. Moreover, the system
adopts the scaffolding theory (Bruner, 1983), which gradually builds their learning patterns. Through this theory, the
system can provide students with information and materials they need. The success of the scaffolding depends on the
precise evaluation of the learning outcomes such that the learning scaffolding can be removed properly. Therefore, a
reliable evaluation system is provided so that learners can determine their progress. As a result, they can set

ISSN 1436-4522 (online) and 1176-3647 (print). © International Forum of Educational Technology & Society (IFETS). The authors and the forum jointly retain the
80
copyright of the articles. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies
are not made or distributed for profit or commercial advantage and that copies bear the full citation on the first page. Copyrights for components of this work owned by
others than IFETS must be honoured. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior
specific permission and/or a fee. Request permissions from the editors at [email protected].
reasonable goals in the system and even develop their SRL skills as they go through each self-regulated learning
cycle.

Another aim of the system is to support a mobile learning environment for learners through modern wireless
technologies via mobile devices. The system enables learners to access learning materials easily and conveniently.
Moreover, the system automatically observes learners' behaviors to help them terminate unhelpful habits, e.g., using
instant messengers or surfing while learning. The mobile learning environment also enables them to share learning
materials, allowing them to adjust their strategies based on the data from their companions. Because of this, the
system would have the characteristics of a mobile learning environment: Urgency of learning need, Initiative of
knowledge acquisition, Mobility of learning setting, Situating of instructional activity, and Integration of
instructional content (Chen, Kao, & Sheu, 2003).

The rest of the paper is organized as follows: Section 2 introduces the design rationale of the proposed system as
well as previous works on SRL. Section 3 introduces the proposed SRL system. Section 4 describes the experimental
results, and Section 5 concludes the study.

Review of Related Literature and Design Rational

Self-Regulated Learning Theory

While there are various explanations and studies that focus on the definition of SRL (Butler & Winne, 1995; Pintrich,
2000), it can be simply described as a learning process with four attributes (Schunk & Zimmerman, 1994):
 Intrinsically or self-motivated: Self-regulated learners tend to maintain learning behavior with a very strong
motivation. Learners can raise this motivation through some practices, such as setting learning goals.
 Planned or automatized: Self-regulated learners are apt to use some strategies along with their learning
processes, including both cognitive and self-regulated strategies. Generally, learners improve their learning
performance when using self-regulated strategies rather than cognitive strategies. Self-regulated strategies
contain goal-setting, goal-planning, organization, transition, exercise, and so on. A self-regulated learner needs
to effectively use self-regulated strategies for his learning.
 Self-aware of performance outcomes: Throughout the learning process, self-regulated learners sharpen their self-
awareness toward their learning behavior. To approach an ideal outcome, self-regulated learners should be
aware of their own learning qualities, and change the behavior or strategies correspondingly.
 Environmentally/socially sensitive and resourceful: The learning environment and resources can affect one's
learning pattern. Self-regulated learners have better skills in seeking learning resources or support. With such
ability, they should arrange the environmental conditions and search for other resources effectively.

Knowing how to possess the above attributes should be considered when designing an adequate system for self-
regulated learners. Once such attributes are possessed, learners can then skillfully self-regulate their learning.

Figure 1. A cyclic model of self-regulatory learning (Zimmerman et al., 1996)

Zimmerman et al. (1996) proposed a self-regulatory learning cycle in order for learners to gain SR skills, as shown in
Figure 1. The cycle involves four interrelated processes which assist learners in evaluating their performance.
Generally, learners carry out their plans by themselves in these processes. Thus, such a model enables students to
arrange their own learning and voluntarily fulfill it at the same time.
81
SRL with the Support of Computer and Wireless Technologies

With the aid of modern technologies, students can learn efficiently and achieve remarkable performance. Unlike in
the traditional face-to-face set-up, today’s students can individually determine when and where to learn. Therefore,
many computer-based systems have been proposed to enhance a person’s performance when he or she learns
individually (Hadwin & Winne, 2001; Dabbagh & Kitsantas, 2004). However, Azevedo, Cromley, Thomas, Deibert,
& Tron (2003) indicated that, when receiving no assistance, students are less effective at regulating their learning in
their hypermedia environment. Because of this, some computer-assisted SRL tools have been proposed (Hadwin &
Winne, 2001; Dabbagh & Kitsantas, 2004).

Hadwin & Winne (2001) proposed a prototype electronic notebook, CoNets2, to support self-regulation through
explicit scaffolding. Its system can support monitoring and controlling engagement in the phases of SRL, but
unskillful self-regulated learners may shun the tool because CoNets2 lacks enough functions to motivate their
learning. Moreover, the tool is often limited to taking down notes, and does not develop important SRL skills such as
goal-setting, scheduling, and self-evaluation.

Dabbagh & Kitsantas (2004) classified Web-based pedagogical tools (WBPT) into four classes: (1) collaborative and
communication tools; (2) content-creation and delivery tools; (3) administrative tools; and (4) assessment tools.
These tools are also examined on their support of self-regulatory attributes. However, an integration of these tools
should be promoted to gradually make learners become skillful in SRL.

Zurita & Nussbaum (2004) proposed a constructivist learning environment, which allows students to build up their
own knowledge. Wireless interconnected handhelds are used in such an environment to achieve the creation of new
knowledge. By using wireless interconnected handhelds, students can able to modify their current knowledge
schemes which could integrate new information and acquire new knowledge. The process of knowledge construction
partially matches the processes of a self-regulatory learning cycle. However, the environment can only be used for
information sharing, and is unable to help students completely monitor their learning strategies and outcomes.

Looi et al., (2009) designed and implemented a software on a mobile device called the 3Rs (reduce, reuse, and
recycle) software. The mobile device used in the activity can lead students to carry out learning tasks in challenge-
experiential cycles, including Challenge, Experience, Reflecting, Planning, and Applying. The activity also partially
matches the self-regulatory learning cycle. However, during the process of Experience, learners may not receive any
assistance, and their learning experience is not recorded and used for future leaning. The comparisons of the
proposed system and the aforementioned tools are summarized in Table 1.

Table 1. Comparisons of the proposed system to the related works

CoNets2 WBPT Constructivist 3Rs software The
(Hadwin & (Dabbagh & Learning Environment (Looi et al., proposed
Winne, Kitsantas, (Zurita & Nussbaum, 2009) system
2001) 2004) 2004)
Support self-regulatory Yes Yes Yes Yes Yes
attributes? (Partial) (Partial) (Partial)
Support scaffolding? Yes No No No Yes
E-learning? Yes Yes Yes Yes Yes
M-learning? No No Yes Yes Yes

The Proposed Learning System

Basic Concept

Shih, Chang, Chen, & Wang (2005) proposed the prototype of the system. According to learner feedback and
opinions from educational theorists, we will continually enhance the proposed system to efficiently improve the SRL
performance of learners. This section introduces the mapping of the proposed system and the cyclic model.
Moreover, the section also tackles the functions and the design considerations of the proposed system.

82
This paper proposes a state transition diagram that indicates the behavior of a learner by using the cyclic model of
SRL to become self-regulatory (as shown in Figure 2). The diagram consists of seven states, indicating the actions of
a learner in SRL. Among these seven, Activity Scheduling, Learning and Monitoring, Learning Evaluation, and
Analysis are the major states that map the four processes in the cyclic model of Zimmerman et al. (1996).

Figure 2. State transition diagram of the system

Through the proposed system, learners in the Activity Scheduling state can obtain information on what to learn and
can arrange suitable times according to the provided information, all of which leads to the Goal Setting and Strategic
Planning process. After setting their schedules, learners enter the Learning and Monitoring states. Because learners
can undertake scheduled activities by using various strategies while they are being observed, the state maps the
Strategic Implementation and Monitoring process. In the Learning Evaluation state, learners can assess their progress
through tests. Accordingly, the state also maps the process of Strategic Outcome Monitoring. In addition, learners
can evaluate their development by means of varied statistical charts, and can discern their SRL patterns in the
Analysis state before it leads to the Self-Evaluation and Monitoring processes. In obtaining a detailed understanding
of their learning characteristics, students can then go into the Activity Scheduling state again, and arrange more rigid
schedules for future learning.

In addition to the aforementioned states, the Synchronization, Help Seeking, and Schedule Reviewing states are also
involved in the system to help learners gradually develop their SRL skills.

Figure 3 illustrates the architecture of the proposed SRL System, which supports the state transition diagram in
Figure 2. The black arrows in Figure 2 are the data flows between subsystems. This means the learning template is
used to help learners in their schedule. The schedule is the learning schedule arranged by the learners. According to
the learning schedule, the proposed system can know when learners plan to learn and observe their behavior. The
result represents the information generated when learners use our system.

Generally, beginners or unskilled learners cannot arrange their learning well because of their lack of experience in
self-regulation. As such, instructors can use the instructor side system to help them in provide directions and scope,
or suitable scaffolds. On the other hand, the learner side system aims to form a pleasant SRL environment wherein
they would be able to practice their SRL skills. The system is planned to be installed in portable learning devices,
where they can schedule, perform, and evaluate their progress anytime and anywhere. In the following, the design of
the functions provided by the proposed system is presented according to the state transition in Figure 2 and the data
flows in Figure 3.
83
 Figure 3. Architecture of the Self-Regulated Learning System

The Data Flow of the SRL System

However, in order to help learners efficiently arrange their learning activities, the information is controlled by a
Scaffold Support Module based on their SRL performance. The purpose of scaffolding is to provide novice learners
with limited complexities of learning context and to remove limits gradually until they become more skillful (Young,
1993). Therefore, the interface initially shows much information until they become more skillful at SRL, so the
learners can then control their learning gradually. Notice that learners are not forced to use given information, but
can decide to refer to the given information.

Figure 4. Learning schedule planning

Previous research have pointed out that intervention increases cognition and motivation, and leads to the
development of self-regulated capabilities (Hofer, Yu, & Pintrich, 1998). In our design, instructors are involved in
helping learners become skillful in SRL through the Content Accessibility Subsystem. Instructors can conveniently
design and give assignments and activities to learners. The subsystem also generates a learning schedule template

84
based on the assignments given by the instructors. Interfaces in Figures 5 and 6 enable instructors to arrange these
assignments and set up the access to learning materials. Through the interface in Figure 5, instructors can set up
detailed information, including degree, session, semester, and credit, in the template. The information can also be
shared by instructors to other instructors, making it easy for them to design templates.

Figure 5. Learning schedule template design

Once learning materials are available, instructors can arrange the details of an assignment through the interface in
Figure 6. Instructors can set up information on a learning unit, including its type (activity, self-examination, or
discussion), suggestion time, and materials. This way, the information can help learners in their schedules and can
also be used by the Scaffold Support Module to generate supplemental information and materials. Similarly,
instructors can query designed learning units, which can be directly imported to a schedule template. After the
instructor's arrangement, the subsystem then generates a schedule template for learners. Through the aid of wireless
technology, the system will automatically download the learning templates from a central database whenever
learners move into Wi-Fi hotspots.

Figure 6 Arrangement for the details of a learning template

85
After scheduling, the system automatically synchronizes to a central database and downloads learning materials
when available. Once learners enter the Schedule Reviewing state via the learning review tool of the Learning
Subsystem, a calendar-like interface, where the scheduled and learned activities are marked, is provided, as shown in
Figure 7. Each selected item on the interface represents a learning activity, and the interface can show basic statistics
(e.g., learning time and the number of interruptions) of a selected item. This information helps learners find their
preferred schedules (e.g., in the afternoon or in the morning).

Figure 7. Learning review tool

When in the Learning and Monitoring states, learners start by clicking the “Learning” button on the learning review
tool. The Learning Subsystem uses two tools, Hyperbook and Hyperpen, to enrich the learning experience.
Hyperbook is a hardcopy book with reference tags. In principle, reading a hardcopy book is comfortable for learners.
However, the content of a book is limited and fixed. In learning devices, staring at the monitor for a long time tire
students easily. Furthermore, they may shun the complicated operation of learning devices because they may need to
alternately use varied input equipment to obtain supplemental materials. Thus, to facilitate operations, students can
use a scanning device, termed Hyperpen, to scan reference tags for more supplemental materials, such as Flash
(Macromedia, 2005), audio, and video. Supplementary materials will then be shown in the learners’ mobile devices.

Hyperpen is embedded with a Bluetooth solution (Bluetooth, 2003) to avoid cables that may distract students as they
scan. Scanned keywords are then submitted to an Internet dictionary, such as Yahoo! Dictionary (Yahoo! Taiwan
Inc., 2007), or the database, such as Answers.com (Answers.com, 2005). Because wireless technologies are heavily
promoted, hotspots can be found all around, and almost all learning devices have WLAN capabilities. Therefore, the
function of searching supplemental materials on the Internet can be carried out everywhere. Figure 8 shows the
system interface where learners use Hyperbook and Hyperpen to avail of supplemental materials on the Internet.
Hyperbook is manufactured by HardSCORM Editor (Wang & Shih, 2006), an authoring tool that conforms to
SCORM 2.0 (ADL Technical Team, 2006). The Content Accessibility Subsystem is able to recognize the edited
courses and can split these into several learning activities based on their metadata.

Because the most important performance control process that distinguishes skillful from naive self-regulated learners
is self-monitoring (Zimmerman & Paulsen, 1995), an Event Monitoring Subsystem is needed to observe their
behavior. If learners can monitor their own progress, their academic performance, achievement, time on task,
classroom behavior, and problem-solving abilities can be improved (Lan, 1998). We therefore give much attention to
recording learner behavior.

86
 Figure 8. Searching the database in Internet (Answers.com, 2005)

Monitoring items include the learner’s schedule and learning behavior. Schedule includes data on whether a learner
studies on time and how much time a learner spends on an activity. Learning behavior involves what students do
while learning, such as the time and reasons of interruptions, and the frequency, quantity, and their ways of seeking
help. Monitoring items can be also easily recorded when learners engage in learning activities. Some learning
interruptions may be caused by various situations, such as the presence of a TV or domestic errands. It is impossible
for a computer-based system to record events automatically. Because the proposed system is designed for students
who want to become self-regulated learners, our system provides an interface for them to manually input
interruptions easily. Monitoring items will be demonstrated in the form of charts after the behavior analysis so
learners can quickly understand their study habits.

Psychologists argue that regularity, referring to the need to observe behavior frequently instead of sporadically, and
proximity, referring to behavior that should be recorded close in time to its occurrence, are important characteristics
of effective self-monitoring (Bandura, 1986). The proposed system provides an easy self-monitoring environment
that has the characteristics of regularity and proximity. For regularity, the Event Monitor Subsystem continuously
observes learners’ behavior anywhere and anytime they experience SRL. As for proximity, learning behavior can be
monitored immediately. Therefore, it is expected that the system can help them understand their learning habits.

The learners need to evaluate their progress either during or after an activity because both the difficulty of the
examination and lack of preparation may incur disappointing performance (Ghatala, Levin, Foorman, & Pressley,
1989). The evaluation should be based on both objective and subjective criteria. The system has a Self-Evaluation
Subsystem that includes an Assessment Module and a Self-checking Module to obtain objective and subjective
cognitions, respectively. Learners are able to identify the gap between what they think about their learning and what
learning outlook they actually have by comparing the objective and subjective cognitions.

After finishing an activity, learners then proceed to the Analysis state, where their behavior is examined. Through the
aid of data mining, meaningful information is dug up from the data in the learners’ profile. The information includes
the differences between learning and the scheduling of an activity and the expected and real scores, among others.
This information is used to determine barriers and bottlenecks in the learning processes, and to find solutions to
certain problems. Students can also read others’ learning analysis, which may help them to either adopt strategies to
other learners or help them find their fulfillment (e.g., rank high among ones classmates.) According to the
monitoring items, many kinds of analysis charts for learning time, interruptions, used materials, and so on are
provided.

When in the Analysis state, learners are able to better understand their learning patterns when provided with the
aforementioned information. Learners therefore try to use various strategies to improve their performance. The
87
instructors can also receive the feedback information for the preparation of following learning templates through
information exchange. In addition, the system automatically downloads other learners’ profiles so that those who
learn the same activity can refer to the learning strategies of others. By entering the Activity Scheduling state again,
they better arrange their schedules based on the experience acquired from prior learning activities, allowing them to
perform more effectively. In the process of doing so, they can be self-regulated learners.

Evaluation of the proposed SRL system

An experiment was conducted in a high school to demonstrate the effectiveness of the proposed SRL system. The
experiment addresses the issues: “Can the system help learners possess the four self-regulatory attributes?” and
“How much does the system help learners to improve their SRL efficiency?”

Method

In the experiment, the target learners were secondary students because we assumed that, though they were eager to
learn, they still lacked SRL skills. The learning topic was English. The experiment consisted of two steps to evaluate
the effectiveness of the proposed system. The first step was to classify the students who learned English in a self-
regulatory way. Students performed SRL without the help of the system in this step. Since the number of students in
this step is large enough, the classification indicates the SRL types of secondary school students in Taiwan. This
classification was used in Step 2 to identify the SRL types of students. Through this step, the following step
discusses the differences between SRL types of students before and after using our SRL system. A pre-test and a
post-test were conducted in the second step to evaluate the improvement of the students’ SRL skills after using the
proposed system. The SRL types of learners in Step 2 were classified into three types based on the classification built
in Step 1. The analysis of the experiment mainly focused on the learners in Step 2 because the proposed system was
involved in their learning. The impacts of the use of the proposed system on the learning performance of the different
types of SRL learners are observed in this step.

In Step 1, four grade ten classes (42 students in one section, and 43 students in the others) were chosen to self-
regulate their English learning for 10 weeks in 2004. Students had the same English instructor and were asked to
study 10 lessons from the Studio Classroom magazines (StudioClassroom, 1962), a popular English instructional
magazine in Taiwan. Their instructor taught them some SRL skills and gave them hardcopy forms to record their
learning behavior and their reflections. After the ten weeks, every student was asked to fill out a MSLQ (Motivated
Strategies for Learning Questionnaire). The original MSLQ was designed by Pintrich, Garcia, & Mckeachie (1993)
to assess college students' motivational orientations and their use of learning strategies. Wu and Chan translated the
MSLQ into Chinese and modified the questionnaire for elementary school students (Wu, 1998). In the experiment,
the MSLQ was used for high school students taking up English. There are 91 items in the MSLQ, and some are listed
in Table 2.

The questionnaire used a seven-point Likert-type scale. The students would receive one to seven marks for each item
in the MSLQ. They were divided into three groups (high SR, medium SR, and low SR) according to normal
distribution. Students whose SRL scores were higher than 485 marks (25% students) formed the group with high SR,
students whose SRL scores were between 484 marks and 423 marks (50% students) formed the group with medium
SR, while the others formed the group with low SR (25% students).

Table 2. Partial items in the MSLQ used in the experiment

I believe I will receive an excellent grade in class.
I'm certain that I can understand the most difficult material in English Learning.
I am very interested in the content area of English learning.

In Step 2, 17 volunteers from one of the grade 11 classes (apart from the four classes in Step 1) were involved in a
three-week SRL. Each student was given a Hyperbook, a Hyperpen, and a tablet PC. The Hyperbook contained six
English lessons from the IVY magazines (Ivy League, 2006), which are also popular English learning materials in
Taiwan.

88
Before the experiment, a pre-test was employed for the 17 students. The MSLQ used in Step 1 was also used in the
pre-test, with the goal of determining the SRL patterns of the students before using our system. Three weeks later,
the students took a post-test. They were asked to fill out two questionnaires. One was an MSLQ, which was used to
discern the SRL patterns of the students after the experiment. The other was the Self-Regulated System Indication
Questionnaire (SRSIQ), which was used to evaluate the support of the self-regulatory attributes.

Generally, an SRL can be surveyed in different psychological dimensions of research by using following scientific
questions: why, how, what, and where (Schunk & Zimmerman, 1994). SRSIQ asked these scientific questions in a
questionnaire filled out by students, as listed in Appendix A. The questionnaire applied a five-point Likert-type scale
ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). The results retrieved from this questionnaire were used to
determine the assistance of the system during the students' SRL.

Results

The experiment results consisted of two parts. The first focuses on the analysis of the four self-regulatory attributes
based on the SRSIQ results of the students. The second part looks into the students’ progress in acquiring the SRL
skills.

The Support of Self-regulatory Attributes

First, the students were asked if the system helped them acquire the four self-regulatory attributes. The reliability of
whether our system motivated students is 0.81, implying that the questionnaire items have a high reliability. The
result shown in Table 3 indicates that the system supports the attribute (Mean=3.329, SD=0.897). This subscale is
greatly influenced by the amount of motivation inspired by our system. The SRL system encourages students to use
supplementary multimedia materials (Q13, Q14, Q15, and Q16) and learning analysis (Q27). Most students thought
that supplementary multimedia materials were useful for their learning. According to the results of the behavior
observation, each student was willing to access the supplementary materials (74.14 materials per student) Therefore,
it can be shown that the supplementary materials and the methods used to access the Internet enriched their learning
experiences and made them eager to learn more. Therefore, the students had spontaneous pleasure in learning
English.

Table 3. Item statistics of attribute: Intrinsically or self-motivated

Mean Std. Deviation N
Q13 3.24 .970 17
Q14 3.47 .943 17
Q15 3.47 1.007 17
Q16 3.47 1.068 17
Q27 3.00 .707 17

The reliability of whether our system helps students become systematic is 0.758. This indicates that the questionnaire
items also have a high reliability. The results also show that all the functions enabled learners to use proper learning
strategies (Mean=3.34, SD=0.749). Among the questionnaire items listed in Table 4, prior scheduling experience
gave significant assistance to students in planning their schedules (Q9). Nevertheless, item Q10 indicates that the
interface of the learning review tool should be more user-friendly. Because the interface of the tool is divided into
two parts, students had to frequently switch from one part to the other in order to view their scheduled and learned
activities. Aside from this, other functions (e.g., synchronizing learning records and starting a learning activity) were
executed through the tool as well, so students thought that the operation of these functions was complicated.
Therefore, students deemed that the interface was not user-friendly enough, and that it should be simplified and
intuitive.

The response on whether the system helped students become self-aware of their performance outcomes also has a
good reliability (Cronbach's Alpha=0.91). Thirteen items tapped on subjective learning performance. The item
statistics, shown in Table 5, show that the information brought to learners gave them a sense of fulfillment
(Mean=3.357, SD=0.766). Through learning analysis, students thought that our system could precisely monitor their
89
behavior and easily record interruptions. Most students also agreed that the Self-Evaluation Subsystem helped them
to conveniently record the cognition to a learning activity, which was used to determine the gap between subjective
and objective learning achievements (Q18). Students also agreed that the learning analysis was helpful (Q25).
However, though learning monitoring was generally regarded as a useful function to understand one's own learning
status (Q19; Q20), the students felt that the function was a little bit hard to use (Q21). This may be attributed to the
fact that the behavior was observed when students used the functions of the proposed system. In the experiment, the
proposed system was installed in the tablet PC, which ordinarily had no input devices such as a keyboard and a
mouse. Unlike desktop PCs, the operation in the tablet PC was more difficult and unfamiliar to the students.
Additionally, the recognition ratio of the Hyperpen was about 89%. Students were not satisfied with the ratio. "It is
hard to scan the tags," one student said. Therefore, our future work should address this because higher recognition
rates can encourage the use of our system.

Table 4. Item statistics of attribute: Planned or automatized

Mean Std. Deviation N Mean
Q1 3.12 .781 17
Q2 3.12 1.166 17
Q3 3.29 .849 17
Q4 3.29 .686 17
Q5 3.47 1.068 17
Q6 3.47 .717 17
Q7 3.47 .874 17
Q8 3.53 .874 17
Q9 3.59 .870 17
Q10 3.06 .966 17
Q24 3.41 .795 17
Q25 3.29 .686 17
Q30 3.29 .772 17

Table 5. Item statistics of attribute: Self-aware of performance outcomes

Mean Std. Deviation N
Q11 3.41 .939 17
Q12 3.35 .702 17
Q18 3.47 .800 17
Q19 3.29 .920 17
Q20 3.35 .702 17
Q21 3.00 .866 17
Q22 3.41 .795 17
Q23 3.41 1.004 17
Q25 3.47 .717 17
Q28 3.24 1.200 17
Q29 3.41 .870 17
Q32 3.47 .800 17
Q33 3.35 .931 17

On average, students were positive on the attribute Environmentally/socially sensitive and resourceful (Cronbach’s
Alpha=0.852). The result in Table 6 indicates that the function of the learning record synchronization (Q31), the
learning materials (videos (Q13), pronunciations (Q14), translations (Q15), and phrases (Q16)), as well as Internet
searching (Q17) could inspire students to obtain and seek useful learning resources, as shown in Table 6. Among
them, Searching on the Internet (Q17) was the least useful function. Because the topic was English, the students
thought that the resources provided by the system were enough and so they had less will to search for extra resources
from the Internet. In the future, the system should be modified to support different types of help seeking functions for
different kinds of learners. For example, the system should enable skillful learners to search for extra materials and
allow unskilled learners to become accustomed to using additional resources.

Genreally, the target students deemed that the proposed system could help them possess the four self-regulatory
attributes, albeit some functions have to be improved.
90
SRL Effectiveness

In this section, the SRL scores of the students are studied. These students are also classified into three groups,
according to the results in Step 1. Of the total number of students, 71% increased their SRL scores after using our
system. On average, the SRL scores of these students increased by 11.06 points.

For a precise analysis, a T-test was used to determine if the proposed system could efficiently improve the learners'
SRL scores. We find that the difference between the means is not statistically significant (t = -1.606, df = 16, p >
0.05, one tailed), so the system cannot help students to significantly improve their SRL performance. This may be
because SRL skills should be developed over a long period of time, and three weeks of SRL may not be long enough
for learners to improve their SRL skills.

Table 6. Item statistics of attribute: Environmentally/ socially sensitive and resourceful

Mean Std. Deviation N
Q31 3.29 .772 17
Q13 3.24 .970 17
Q14 3.47 .943 17
Q15 3.47 1.007 17
Q16 3.47 1.068 17
Q17 3.18 1.131 17

The SRL scores of all students in the group of low SR increased, except for one student. Those who lack learning
experience can easily follow different learning styles, such as computer-assisted learning or distance learning. On the
other hand, skillful learners have more difficulty in changing the learning patterns they have developed (Morgan,
Dingsdag, & Saenger, 1998). Therefore, whether or not the proposed system significantly improves the SRL scores
of the students in the group of low SR is analyzed. The difference between the means is significant at the 0.05 level
(t =-3.136, df = 9, one tailed). The result shows that the SR skills of students in the group of Low SR significantly
improved, which corresponds with the findings of Morgan et al.

Conclusions
This paper proposes an SRL system that involves the self-regulatory learning cyclic and scaffolding theories to
cultivate self-regulated learners. The system aims to construct a mobile, portable, and personalized learning
environment for SRL that can be used anywhere and anytime. To help learners gradually develop their SRL skills,
instructors are involved in the system. The learner side system enables students to start SRL anytime and anywhere,
obtain learning materials and assistance instantly, realize their learning patterns, cultivate their SRL behavior, and
sustain their interest in self-learning.

Generally, the experiment results show that the system has improved SRL skills, though the improvement is not
significant. However, the result showed that the SR skills of students in the Low SR group improved significantly.
Moreover, most of students deemed that the overall interfaces of the proposed system were user-friendly and could
give them valuable progress in SRL. The students also agreed that the system enabled them to acquire the four SRL
attributes.

Based on the suggestions of the students, we will make the interfaces of the SRL system more user-friendly and
improve the recognition ratio of the Hyperpen to facilitate its operations. We also aim to use other devices such as
PDAs or mobile phones as learning instruments because they are easy to carry and can facilitate quick learning. As a
result, learners would more be likely to use the proposed system. Experiments on a larger number of students over a
longer period of time shall also be conducted to improve the system.

Acknowledgement
The work was partially supported by the National Science Council of the Republic of China under Grants NSC 97-
3114-E-119-001, NSC 97-2221-E-032-021, and NSC 98-2218-E-254 -001.
91
References
ADL Technical Team (2006). SCORM 2004 3rd Edition, retrieved Jun. 18, 2009 from http://www.adlnet.gov/scorm/index.cfm.
Answers.com (2005) Answers.com- Online Encyclopedia, Thesaurus, Dictionary definitions and more, retrieved Aug. 23, 2009
from http://www.answers.com/.
Azevedo, R., Cromley, J. G., Thomas, L., Deibert, D., & Tron, M. (2003). Online process scaffolding and students' self-regulated
learning with hypermedia. Annual Meeting of the American Educational Research Association, 31.
Bandura, A. (1986). Social Foundations of Thought and Action, Englewood Cliffs, NJ: Prentice-Hall.
Bluetooth® SIG, Inc. (2003). Bluetooth Core Specification Version 2.1 + EDR, retrieved Jun. 18, 2009 from
https://www.bluetooth.org/spec/.
Bruner, J. (1983). Child's talk. Learning to Use Language, New York: W. W. Norton.
Butler, D. L. & Winne, P. H. (1995). Feedback and self-regulated learning: a theoretical synthesis. Review of Educational
Research, 65(3), 245-281.
Chen, Y.-S, Kao, T.-C., & Sheu, J.-P. (2003). A Mobile Learning System for Scaffolding Bird Watching Learning. Journal of
Computer Assisted Learning, 19(3), 347-359.
Dabbagh, N., & Kitsantas, A. (2004). Supporting self-regulation in student-centered web-based learning environments.
International Journal on E-Learning, 3(1), 40-47.
Ghatala, E. S., Levin, J. R., Foorman, B. R., & Pressley, M. (1989). Improving children's regulation of their reading prep time.
Contemporary Educational Psychology, 14, 49-66.
Hadwin, A. F., & Winne, P. H. (2001). CoNoteS2: A software tool for promoting self-regulation, Educational Research &
Evaluation, 7, 313-334.
Hofer, B. K., Yu, S. L., & Pintrich, P. R. (1998). Teaching College Students to Be Self-Regulated Learners. In Schunk, D. H. &
Zimmerman B. J. (Eds.) Self-regulated learning- from teaching to self-reflective practice (pp. 57–85), New York: Guilford.
Ivy League (2006). Ivy League Analytical English, retrieved May 3, 2009 from http://www.ivy.com.tw/.
Lan, W. Y. (1998). Teaching self-monitoring skills in statistics. In Schunk, D. H. & Zimmerman B. J. (Eds.) Self-regulated
learning- from teaching to self-reflective practice (pp. 86–105), New York: Guilford.
Looi, C.-K., Seow, P., Zhang, B. H., So, H.-J., Chen, W., Wong, L.-H. (2009). Leveraging mobile technology for sustainable
seamless learning: a research agenda. British Journal of Educational Technology, 41(2), 154-169.
Pintrich, P. R. (2000). The role of goal orientation in self-regulated learning. In Boekaerts, M. & Pintrich, P. R. (Eds.) Handbook
of Self-Regulated (pp. 13-39), San Diego: Academic.
Pintrich, P. R., Garcia, T. & Mckeachie, W. J. (1993). Reliability and Predictive Validity of the Motivated Strategies for Learning
Questionnaire (MSLQ). Educational and Psychological Measurement, 53(3), 801-813.
Macromedia Inc. (2005). Macromedia Flash MX 2004, retrieved Sep. 25, 2009 from http://www.macromedia.com.
Morgan, C. J., Dingsdag, D. & Saenger, H. (1998). Learning strategies for distance learners: Do they help. International Journal
Distance Education, 19, 142-156.
Schunk, D. H. & Zimmerman B. J. (1994). Self-Regulation of learning and performance: Issues and educational applications,
Hillsdale: New Jersey: Lawrence Erlbaum.
Shih, K.-P., Chang C.-Y., Chen, H.-C., & Wang, S.-S. (2005). A Self-Regulated Learning System with Scaffolding Support for
Self-Regulated e/m-Learning. Proceedings of the 3rd IEEE International Conference on Information Technology: Research and
Education (ITRE 2005), 30-34.
StudioClassroom (1962). StudioClassroom.com-Your Friend for Life, retrieved Sep. 25, 2009 from http://studioclassroom.com/.
Wang, T.-H. & Shih, T. K. (2006). Integration of multimodal multimedia devices and hardcopy textbooks for supporting
pervasive e-learning. Paper presented at the First International Symposium on Pervasive Computing and Applications (SPCA06),
August 3-5, 2006, Urumchi, Xinjiang, China.
Wu, J.-J. (1998). Consortium of Research on Creativity and Innovation, retrieved Oct. 25, 2009 from
http://tim.nccu.edu.tw/croci/group/wujingji.htm.
Yahoo! Taiwan Inc. (2007). Yahoo! Dictionary, retrieved Sep. 12, 2009 from http://tw.dictionary.yahoo.com/.
Young, M.F. (1993). Instructional design for situated learning. Educational Technology Research & Development 41(1): 43-58.
Zimmerman, B. J., Bonner, S. & Kovach, R. (1996). Developing Self-Regulated Learners: Beyond Achievement to Self-Efficacy,
Washington, DC: APA.
Zimmerman, B. J. & Paulsen, A. S. (1995). Self-monitoring during collegiate studying: An invaluable tool for academic self-
regulation. New directions for teaching and learning, 1995(63), 13-27.
Zurita, G. & Nussbaum, M. (2004). A constructivist mobile learning environment supported by a wireless handheld network.
Journal of Computer Assisted Learning, 20(4), 235-243.

92
Appendix A: Self-Regulated System Indication Questionnaire (SRSIQ)
Q1 Learning Scheduler Subsystem provides clear-cut lesson information to schedule learning.
Q2 With drag-and-drop, I plan my schedule smoothly.
Q3 Learning scheduler subsystem helps me determine the study plan.
Q4 Functions of Learning Scheduler Subsystem are varied and handy.
Q5 Learning Scheduler Subsystem is helpful in setting appropriate learning goals and plan setting.
Q6 Learning Review Tool helps me manage the learning activities.
Q7 Learning Review Tool makes me understand the time spent on prior learning activities.
Q8 I can see every learning activity by examining the information in the Learning Review Tool.
Q9 Tracking the previous learning schedules, I set better plan in the next step.
Q10 Learning Review Tool is designed with user-friendly interface.
Q11 The "pause" function helps me unhurriedly note down every interruption while learning.
Q12 Recording learning time helps me know my own learning progress.
Q13 Introductory videos provided by the system are useful for the beginning of my learning.
Q14 Recordings provided by the system are useful for my learning.
Q15 The function of text translation is useful for my learning.
Q16 The provided vocabularies and phrasal verbs in the Hyperbook are useful for my learning.
Q17 Hyperpen is convenient for searching resources on the Internet.
Q18 Self-evaluation helps me immediately note down my experience and feeling along the learning.
Q19 Items of learning recording and learning monitoring are listed comprehensively.
Q20 Learning and monitoring tools enable me to precisely manage my learning.
Q21 Learning and monitoring tools are designed with user-friendly interface.
Q22 Learning analysis outcome in diagrams and illustrations offers clear ideas.
Q23 Interruption analysis helps me to know where distractions come from.
Q24 In order to keep interruptions out of the process, I amend my learning strategies according to the reasons
of interruptions.
Q25 The statistics coming from the use of provided multimedia helps me know my own learning habits.
Q26 The analysis of learning outcome is a good reference for following a schedule design.
Q27 The analysis of all learners improves my learning desire.
Q28 Online analytical dictionary shows me unfamiliar words.
Q29 The analysis of online resource searching record shows me unfamiliar territory.
Q30 The learning analysis of all learners helps me amend my learning strategies.
Q31 The synchronization of learning records is designed with user-friendly interface.
Q32 The analytical tool assists me in understanding my own learning.
Q33 The analytical tool is designed with user-friendly interface.

93