Editor’s Note: This is
a detailed and exhaustive review or research and praxis for model online courses
based on Data Structures. Interactive multi-media facilitates learner participation
for effective learning. Many options are possible in a media rich environment.
Practical considerations are discussed in detail, and the concepts and implementation
strategies are well worth very close study.
Developing an Online Data Structures Course
Using Authorware
S. Junaidu and J. Al-Ghamdi
ABSTRACT
The potential pedagogical
benefits of online learning are being recognized widely. This is reflected in
the widespread development, both in corporate establishments and in academia,
of remote labs and academic courses. However, a common feature of online courses
today is that they are either in the form of still text only, or interspersed
with some applet animations or incorporate some form of video clips that are
often of low resolutions and which do not lend themselves to easy navigation
by the online learners.
This paper presents our
experience of developing a multimedia-rich, learner-friendly online data structures
course using Macromedia Authorware, the premier development tool used for developing
interactive, media rich solutions for online learning. We note that graphics
animations and active learner participation are crucial components for effective
online learning. We report how our online course carefully addresses these issues
to good effect. We also report results of our students' questionnaires during
the first offering of the course. The results show progressive students' satisfaction
with the course from the beginning to the end of the semester.
1. INTRODUCTION
The hottest development
in using technology for education is the use of the WWW for delivery of course
material and student-tutor interaction and there are countless examples of such
initiatives world-wide [Sme97a]. The World Lecture Hall [Wor96] is an online
repository or listing of such courses; and within the "Computer Science"
classification there are already pointers to 109 courses in computer science
on offer in various Universities, and computer science is just one of 98 such
categories! Most of these courses deliver teaching material organized as electronic
textbook or as electronic course notes that the student can "read"
at his/her own pace. Some also provide multiple routes through the material,
search facilities, integration of text, image, video and interaction, and question-answer
tests. We note that none of the six courses on data structures on World Lecture
Hall site provides more than the syllabus and electronic form of the course
material.
A common limitation of
the online course materials we have reached is that they lack a user-friendly
environment that an online learner requires. Whereas our results of students
surveys indicate that users are in continuous demand for more control on how
they interact with the online course material. Our online course attempts to
provide a modest solution to this demand by making it possible for our online
course users to
- disable audio in the presentation and view it in animation and text mode
- pause and resume at any point within the presentation
- navigate, using hyperlinks, through a lecture by sub-topic
We note that students
value these controls highly because they enable progression through the course
material as fast or as slow as the student wants to go and in an order or sequence
driven by the student's own capability and desire. Subsequent offering of our
online course after incorporating these user-control regimes increased the course
rating by 30%. We are currently investigating how to effectively incorporate
even easier navigation mechanisms using sliders and an hierarchy tree along
the lines of File Explorer in personal computers.
Our online course project
is based on Data Structures. This course is the last of a series of three basic
courses (introductory courses according to [CC2001]) offered in the Department
of Information and Computer Science. While this course was more challenging
to start with, we nevertheless took it first for a number of reasons. First,
it was the only course among the three that has not been offered in our transition
from imperative to object oriented paradigm. It will require more work to revise,
decide the content and then implement it online. Secondly, students taking the
course are already quite familiar with the Internet and so have no learning
curve to climb in using the delivery tools. Also, these students already had
a computing culture with the use of PCs being an integral part of their daily
lives (Our survey shows 98% of the students own PCs).
The remainder of this paper
is organized as follows. In the next section we outline the course content development
process. Section 3 discusses the issues of animation, recording and synchronization.
Section 4 presents the design of our presentation template using Authorware.
Section 5 presents how the course material is uploaded onto WebCT and in Section
6 we present our course delivery method. In Section 7 we discuss results of
our students' surveys. Finally, we summarize in Section 8 and conclude with
acknowledgements in Section 9.
2. CONTENTS DEVELOPMENT
The success of this course
is, to a large extent, attributable to the organized and systematic way in which
the course content is developed. The course content is developed continuously
in a cycle of preparation-presentation-feedback-preparation before the material
is in its final form for transferring to the presentation template.
Right at the start of the
course development, we planned that each project member will present to the
group, in each weekly meeting, two-lectures worth of his prepared content. Each
presenter receives detailed comments on various aspects of his presentation
including content, choice of examples, contents organization and nature of animations
suggested. The overall flow of the presentations is also checked to ensure smooth
transition from the lectures prepared by one instructor to another. We also
ensure that subsequent lectures build on examples in earlier lectures as much
as possible.
One of the important requirements of an online course is engaging the learner
in an active way. In an attempt to achieve this, our lecture units contain an
average of two pop-up interactive quizzes that seek to test learners' comprehension
of the content. The quizzes were developed using Authorware's knowledge objects
tool. Various kinds of questions can be created using this tool to
minimize the tendency to feel `lonely' that is characteristic of an online course.
Review exercises at the
end of each lecture material are designed and tailored to re-enforce students
understanding of the relevant material. The laboratory material is developed
hand-in-hand with and compliments the lecture material. As the laboratory session
is the most important component in the course with respect to skills building,
homework questions are designed to enable students to build on the laboratory
coverage. This provides the potential for concretizing the students’ skills
and building their confidence for independent self-study.
From the foregoing discussion,
it is clear that the time investment during content development is high. Content
development, voice transcript writing and content transfer to the presentation
template amount to about 40% of the online course development time. We believe
that this investment is worthwhile in order to produce duplication-free, tightly
coupled lectures, labs and homework contents. An immediate consequence of this
is the instructors' ability to present, within the same coverage time, and without
overwhelming learners, about 25% more material in the online course than in
a traditional face-to-face course offering.
3. ANIMATIONS, RECORDING AND SYNCHRONIZATION
One of the most important
aspects of an online course is that it should be illustrative. This point cannot
be overemphasized. An online course should use multimedia elements (animations,
sound, graphics, color etc) in a measured way to illustrate and explain important
concepts. As mentioned earlier, animation decisions (what to animate and the
nature of the animations) are best done at content preparation time.
Multimedia allows a variety
of learning style preferences to be accommodated. Learners can choose to study
course contents in a way that suits their preferences with the opportunity to
control their pace. With judicious use
of multimedia, an online course author can really appeal to learners intuition
and, potentially, do better than what can be done in the traditional face-to-face
method of instruction.
Using Authorware's animations
capability we are able to animate algorithms on trees, graphs, hashing and data
compression, amongst others. It is instructive to note that algorithms, which
are otherwise cumbersome to teach effectively in a single traditional lecture
session, have been successfully animated and effectively packaged to be learned
in a fraction of the traditional face-to-face coverage time! Not surprisingly,
students’ performances at examinations demonstrate good mastery of such animated
algorithms.
A major cost of animations
is development time. We note that examples like AVL tree operations, Dijsktra's
shortest-path computation and Huffman coding can each take a whole day to animate
well. This is partly because Authorware gives full control on animations so
that, for an animation element to stand alone, it should be in a separate icon
and its movements and transition effect must be set individually. Although animations
could be incorporated into Authorware presentations from external sources, the
fine granularity of animation elements may not be possible especially since
these animation elements may have to be synchronized with voice elements. In
our development experience, animations account for about 30% of production time.
If an online course is
to include voice, then the issues of recording software, recording format, recording
environment and synchronization have to be carefully considered. After trying
a few voice editors we settled for Sound Forge, from Sonic Foundry, as our recording
software. The choice of a recording format is a trade-off between quality and
space requirement. We found that a 16-bit, mono wav file at 98 KHz saved in
PCM format produces acceptable quality sounds. Recording should be conducted
in an environment in which background noise has no effect on sound quality.
Synchronizing multimedia
components must be done carefully. Animation and sound require a certain length
of time to play-time that can vary from computer to computer [Macr01]. It is
crucial to make sure that these components start and stop at the right time
when they are played together. Otherwise the presentation will not be smooth.
Authorware provides two functions SyncPoint() and
SyncWait() to synchronize sounds with
display text and graphics. Synchronization using these functions usually requires
repeated testing to determine how long each component takes to load and play.
Where any noticeable pauses occur when a component is loaded into memory, the
start and stop points must be adjusted for each component.
Our synchronization alternative
to using the above functions is to break our sound files into smaller units
as dictated by the corresponding text and/or graphics that the sound segments
explain. The associated text/graphics and sound segment are then played concurrently
or sequentially as desired. This provides an additional low-level synchronization
control and makes the presentation more susceptible to future enhancements.
Recording, synchronization
and packaging our course material for delivery onto the Web accounts for about
25% of the course development time.
4. PRESENTATION TEMPLATE DESIGN
Before presenting packaging
and delivery to the Web, we highlight the issue of designing a presentation
template. There are two major issues here: the presentation template itself-its
size, color and font styles and size-and the navigation options to be implemented.
When designing a presentation
template, the developers should consider carefully the kinds of machine that
their clients will be using in order to arrive at a correct size. A template
size of 640 by 480 is likely to cater for most computer screens in use today
and we use it as a good common denominator. A sample page of our presentation
template is shown in the appendix.
We chose and implemented
standard navigation options found in typical GUI-based systems. Each subtitle
in the front page of a lecture is a link that leads to the first page on which
discussion of that point begins. At the bottom of the presentation template
are different navigational buttons and sound/presentation control buttons. The
navigational buttons enable the students to move to the next/previous page,
move to the next/previous section, repeat the current page, move to the beginning
page of the lecture and to search for a particular phrase in the lecture.
From our experience in
this course, there is demand from students to have more control on navigation
through the material. We are considering adding inter-lecture links and a slider
to give additional control on navigation. A sample page of our presentation
template is shown in the appendix.
5. UPLOADING ONTO WEBCT
At this point all the
technical issues of course development have been completed. We are now ready
to compile and upload the material onto the Web.
Before uploading to the Web the wav files produced during the recording process
should be compressed for increased efficiency. Compression and conversion
formats may depend on the authoring tool used. Working with Authorware, we converted
our wav files to Shockwave format. Shockwave Audio is a technology
that makes sounds smaller and plays them faster off disk or over the Internet.
Shockwave Audio can compress sounds up to a ratio of 176-to-1. It also makes
them streamable - meaning that they can begin playing before they're completely
loaded into memory.
The streaming technology
of Authorware enables a presentation piece to be packaged especially for the
Web so that it can be efficiently downloaded over the intranet or Internet from
a compatible browser. Streaming consists of a packager and a player. The packager
prepares the piece and breaks it into segments for faster transmission over
the network. The packager also creates a map file that tells the web player
what to download, when to download it, and where to put the downloaded segments.
The web player manages the downloading according to the map file and runs the
piece independent of the Authorware application.
After packaging the course,
it is uploaded onto the Web under the WebCT course delivery tool. WebCT contains
a rich set of tools for managing communication, student tracking, assessment
and tools for conducting and analyzing surveys. Because an online course requires
periodic review and enhancements, we take advantage of these tools to monitor
the course for future improvements.
6. COURSE DELIVERY
As we mentioned earlier,
our online course goes beyond the common provision of many online courses of
mere electronic form of the course material. Our course provides much richer
course material that is both user-friendly and multimedia intensive. Although
the course could be offered as a replacement for lectures in the traditional
brick-and-mortar teaching environment, it is currently offered as a compliment
to the latter. The course is intended to replace the traditional face-to-face
delivery method in the near future. The traditional three-hour weekly teaching
times are replaced with a single weekly tutorial hour. Being a four-credit course,
the laboratory component is maintained as a traditional face-to-face weekly
meeting with the students.
The tutorial sessions
are blended into our delivery method to increase personal contacts between the
students and the instructors. It is important for us to address the social consequences
of students not attending class as part of a large group and we do this with
group tutorial sessions that provide an opportunity for socializing, brainstorming
and chance conversations. In these weekly sessions students meet the facilitator
to discuss and clarify course material for the first twenty minutes of the hour.
The remaining part of the hour is devoted to either a quiz or discussion of
the quiz taken during the last tutorial session. Students' tasks alternate between
taking a quiz and homework submission during the weekly face-to-face sessions.
The laboratory face-to-face
meeting is one of the most important components of the course especially in
building the students' skills. The laboratory tasks are more or less a do-it-yourself
exercise for all but the below-average students. The laboratory documents, like
the course material, are carefully written to reflect and reinforce the understanding
of the course material. The laboratory sessions are often started with short
quizzes on the material to be covered in the current laboratory session. The
reinforcement of material in the tutorial and the laboratory sessions leads
to a broader skills base with students taking more control of their learning
activities. In our mode of delivery, tutorials assume a more important and central
role than otherwise as they become the major point of contact among students
and between students and lecturer.
The obvious benefits that
a series of online lectures have are that students take control of their own
learning and do so at their own rate, albeit moderated by the facilitator in
charge of the course. The passive transfer of information as in conventional
large-class lectures is replaced by personalized delivery, with the student
determining when, what, how fast, how much and how often material is covered.
The content of the course
is organized into forty lectures as would typically be presented in the traditional
face-to-face lecture method. The content of the course has been gathered into
lecture-sized units for easier navigation. Each lecture takes, on average, 35
minutes to complete when viewed with the full audio explanations and animations.
As a supplement, the students
are also provided with paper copies of the screen dumps of the Authorware presentation
for off-line study. The students make use of the hard copy to write additional
notes that they transcribe from the online presentation. The course material
is uploaded onto WebCT is password protected and each student is provided with
an account with which to enter the course page. A CD is also provided containing
the compiled course material as posted on WebCT. For online study of the material
in the college laboratories, students are provided with walkman-style headphones
so as to reduce disruption in the laboratories caused by audio over speakers.
7. STUDENTS FEEDBACK
Following Smeaton [Smea97a],
we evaluate our online course according the following three criteria:
- Students' performance
- Online course system usability
- Scale of investment needed to create and maintain the online course.
The performance of students
is a measure of how well they have learned from the online versus the traditional
mode of lecture delivery. Since performance in examination is not an adequate
measure of comparative students performance (because of varied learning capabilities
of students from one semester to another, relative difficulty of examinations,
etc.), we relied mainly on our teaching experience and feedback from instructors
teaching courses that require this online course as prerequisite. Experience
of instructors who taught this course earlier in the traditional face-to-face
model found that online students demonstrate a much deeper understanding of
the course material. This is partly due to the relatively more active role that
students play during online learning. Furthermore, feedback from instructors
teaching advanced level courses showed an increased satisfaction in the level
of students’ performance as a result of their background in the basic courses.
We analyze the usability
of our online course system using the 5 criteria due to Nielsen, [Nie193]:
1. Learnability. This refers to the user's ability to quickly learn
how to use a system. We base our analysis of our system's learnability based
on the two questionnaires given within each semester. In each of these questionnaires,
about 80% of the students say they find the system easy to use. From our surveys,
however, there are requests for additional control on navigation, which we
are trying to incorporate.
2. Efficiency of use in our online course can be assessed in terms
of the effectiveness of our animations. Our experience with teaching in the
face-to-face method is that it is difficult to effectively teach topics like
AVL rotation cases and the Dijkstra's shortest path algorithm using reasonable
examples each within a lecture session. With our carefully designed animations,
however, we are able to demonstrate these algorithms in about ten minutes
each. Furthermore, a student can replay these animations to improve comprehension,
unlike in the face-to-face method where the instructor erases such illustrations.
This is one point where the online delivery method has a clear edge over the
face-to-face method.
3. Memorability is a measure of student's ability to grasp material
quickly. Statistics from results of our questionnaires indicate that most
students grasp the algorithms being illustrated after following our presentation
slides the first time. Normally, they revisit the presentations to refresh
their memories about additional concepts explained in the audio parts of the
presentations.
4. Errors refer to the number of minor corrections (typos, etc.) and
major problems (system downtime, system failures). Admittedly, typographical
errors and minor formatting errors are some of the main things we had to face
during our first revision of the course material. We maintain log files of
errors we find as well as those received from user feedback.
5. Subjective satisfaction with the online lectures will be measured
using post-course questionnaires. There are currently ten students who took
the online course and are giving us feedback and who are working part-time
on developing other online courses. These students are also a source of valuable
feedback from their friends currently taking the course. We discuss these
and other feedback we receive through the chat tool of WebCT in our regular
meetings. Critical feedback is acted upon immediately while other concerns
may be left for the next review period.
Overall, results of our
students’ survey of the online course are encouraging. Surveys are conducted
at the start, in the middle and at the end of the semester. The results indicate
increased students' satisfaction through the semester. Early in the semester,
about 60% of the students indicate that they were worried and about 30% say
they oppose the methodology. In the last set of questionnaires, however, about
60% of the students say they are extremely satisfied, 30% moderately satisfied
and about 5% said they were not satisfied at all.
The final aspect of the
evaluation of our work concerns the scale of investment needed to create and
maintain online lectures. As mentioned earlier, the course material is hosted
on WebCT for which there was an existing administrator before our online course.
The WebCT administrator is responsible for creating students accounts and troubleshooting
the system.
The big investment in
creating the online lectures has been the time required. Six instructors worked
almost round-the-clock for about three months to produce the first online course.
The same instructors worked part-time for three months in the first review period
of the course. There are now ten students working part-time on the revised course
and comparing it with the version they studied. The students are also exploring
the authoring tool, Authorware from Macromedia, with the view to help the instructors
cut development time for subsequent online courses.
Other costs include the
cost of the Authoring tool, cost of headphones for students and cost of backup
hardware. Although each of the course instructors meets the students once in
a week, the instructors put significant amount of effort in the online course
delivery. In addition to frequent quizzes and homework, the instructors have
to respond to students’ questions online or through WebCT's chatting tool as
well as conducting and analyzing questionnaires.
In spite of the above
costs, online course offerings have, potentially, a number of benefits. First,
an online course provides opportunities for training students to be more independent
learners, better time managers and ultimately able researchers. Online courses
also lead to significant reduction in classroom space usage since the face-to-face
meeting time is reduced from three times in a week to only one. Another attractive
feature of online courses is the potential of being scalable at relatively low
costs. Another important factor is fame; the authors' university is becoming
a pace-setter in this important area of online education. Many other universities
in the Kingdom and the region at large are showing interest in what has been
learned already from this experience.
8. SUMMARY AND FUTURE DIRECTIONS
We have presented the
design and implementation of a media-rich, flexible online data structures course,
which, to a large extent is unique in its kind. This is partly because the course
has gone beyond the traditional online courses that typically consist mainly
of still text presentations. Our online course also supports learner interactivity
and cater appropriately to the needs of the fast as well as the slow learners;
the user can pause, disable voice explanations, repeat a certain point and jump
within a lecture as and when needed.
We have also discussed
the important role that animations play in an online course. We described how
we created effective, reusable animations that are fundamental to illustrate
important concepts. Each lecture unit consists of two short pop-up quizzes,
on average, that play the double role of making the learner participate actively
in the learning process as well as test learner comprehension of the content.
We have also reported in
this paper results of students surveys on the usability of our system and the
effectiveness of the animations and navigation options. The overall results
are very encouraging as they indicate increased students reception of the course
through the semester.
As with every online course
development effort, we plan to put in place a continuous review regime of improving
the course content. Before the next offering of the course we plan to interchange
placement of some materials, incorporate additional solved examples and additional
drill questions. We also plan to add a navigational slider and a keywords search
facility as additional forms of user navigation and control.
9. ACKNOWLEDGEMENT
This research was carried
out under the auspices of a University project in the King Fahd University of
Petroleum & Minerals. We acknowledge the foresight of the University authority
for funding such projects.
We also acknowledge Bruno
Press, author of Data Structures
and Algorithms in Java using Object Oriented Design Patterns. We
made extensive use of the idea of presenting data structures in a unified way
as facilitated through the use of design patterns as espoused in this book.
Finally, we acknowledge
the efforts of our colleagues in the Online Project with whom we have shared
ideas and whose other engagements could not enable them partake in the writing
of this paper.
REFERENCES
1. Smeaton, AF and Crimmins, F (1997a), Virtual Lectures for online Lectures:
Delivery using ReadAudio and the WWW, in Proc. Ed-Media/Ed-Telecom,
Calgary, Canada, 1997.
2. World Lecture Hall,
http://www.utexas.edu/world/lecture,
1996.
3. Chang, C et al. (2001), Computing Curricula 2001 (Computer Science),
Final Report, December 15, 2001.
4. Online Journal of Distance Learning Administration, http://www.westga.edu/
distance/jmainl1.htm1
5. How to Develop an Online Course, http://stylusinc.com/online_course/tutorial/process.htm
6. Using Authorware, Macromedia, Inc., 600 Townsend St.
San Francisco, CA 94103, First Edition, August 2001.
APPENDIX
Collisions
and Collision Resolution Schemes
|
+ Introduction
to Collisions
|
|
+ Collision Resolution
Techniques
|
|
+ Separate Chaining
|
|
Quiz 1
|
|
-- Open Addressing
|
|
Linear Probing (Linear Search)
|
|
Quadratic Probing
(Non-linear Search)
|
|
Rehashing (Double
Flashing)
|
|
Random Increments/Decrements
|
|
+ Quiz 2
|
|
+ Review Exercises
|
|
|
f3 sound i f pause
page 1 of 15
|
Figure 1: Sample Presentation Template Page
About the Authors
Sahalu
Junaidu is
an Assistant Professor of Computer Science. He received his Ph.D. from St. Andrews
University, Scotland in 1998. His areas of interest include parallel computing,
computer networks and e-Learning. He has been a faculty in the Information and
Computer Science Department of King Fahd University of Petroleum & Minerals
since 1999. He has been co-leading an online course development team for the
last 14 months.
Contact Sahalu at sahalu@ccse.kfupm.edu.sa
Jarallah AlGhamdi is the Dean of the College of Computer
Sciences and Engineering at King Fahd University of Petroleum & Menials.
He received his Ph.D. in Computer Science from Arizona State University in 1994.
He worked in research in software engineering and in particular in software
metrics. He worked in computer science curriculum development and is working
in e-Learning extensively in the past two years.
Contact Jarallah at jaralla@ccse.kfupm.edu.sa.
|