REQUIREMENTS FOR EDUCATIONAL SUPPORT TOOLS
IN VIRTUAL WORLDS
Ishbel M. M. Duncan
School of Computer Science, University of St Andrews, North Haugh, St Andrews, Fife, KY16 9SX, U.K.
Natalie J. Coull
Department of Computing, University of Abertay, Dundee, DD1, U.K.
Keywords: Requirements, Evaluation, Educational Tools, Virtual Worlds.
Abstract: Virtual Worlds have been used for online gaming, socialising, business advertising and education. As the
educational uses become more sophisticated from early information advertising and teaching resources to
simulated laboratory and scenarios, it is necessary to determine requirements for tools and virtual systems to
achieve the optimal support possible. This paper discusses the current educational uses of Virtual Worlds
and applies this to a set of support requirements derived for an evaluated support tool for learning to
program. The result is a layered approach, or checklist, to support learning and evaluation for effective and
adaptive online educational support tools including virtual world educational systems.
1 INTRODUCTION
Virtual Worlds (VW) may have started as a
mechanism for young people to enjoy remote
socialising or multiple online gaming, but much
recent effort has gone into developing online
advertising islands for major commercial companies
as well as sophisticated techniques for determining,
and exploiting, youth culture and fashion.
Academics have previously used VLEs to enable
teaching and learning but recently, there has been
interest in using VWs to situate teaching resources,
lectures, online practical sessions, discussions and
general information for students.
Virtual Worlds form part of the domain of
multiplayer online games, without necessarily the
game play. A VW could be entered on an off-line
personal computer if the game world does not
include interaction with other players, such as in the
Sims (http://thesims.ea.com). However, a Multi-
User Virtual World (MUVE) such as Second Life
implies on-line interaction with other users. MUVEs
allow multiple users to access 3D environments at
the same time from different geographical locations.
The educational usage of MUVEs has recently
become an area of research (Getchell et al., 2006,
Salmon, 2009, Edirisingha et al., 2009, Oliver and
Carr, 2009, Duncan et al., 2012) with a large
emphasis on Second Life (www.secondlife.com
). In
2010 over 750,000 unique users spent over 105
million hours online with Second Life (SL). It is a
major supporter of education with over 700
international educational institutions residing within
its islands, or server resource centres. Separately,
use of the Internet has grown to over 2,267 Million
users worldwide in December 2011, 32.7% of the
world’s population. The biggest online populations
are in Asia, 1,016 Million, and Europe at 500
Million users. The biggest penetration of users is in
North America with 78.6% of its population active
online at 273 Million out of 347 Million users
(www.internetworldstats.com
). SL penetration is
therefore small at less than 1%. However, any usage
for educational benefit has a potentially large
audience.
A Virtual Learning Environment (VLE) is
commonly used as a virtual environment for
supporting educational documents, uploading
assessments and grades as well as supporting
communication between students and lecturers or
tutors with a chat mechanism (Jenkins et al., 2005).
Whereas VLEs are dedicated to educational support,
VW and MUVEs are still commonly used for
gaming and or socialising via the use of avatars.
141
M. M. Duncan I. and J. Coull N..
REQUIREMENTS FOR EDUCATIONAL SUPPORT TOOLS IN VIRTUAL WORLDS.
DOI: 10.5220/0003904701410145
In Proceedings of the 4th International Conference on Computer Supported Education (CSEDU-2012), pages 141-145
ISBN: 978-989-8565-06-8
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
They too may support chat communication and can
further the student experience by visualising or
modelling real or imagined world events, scenarios,
buildings or sites. For example, Second Life has
islands dedicated to Higher Education Institutions
which support simulated archaeological digs,
architectural building development, language
teaching and algorithm display.
2 CURRENT STATE OF THE ART
Many articles have been written on the different
aspects of the educational uses of MUVEs (Akpan
and Brooks, 2005, Childs, 2007, Dawley, 2009,
Mason, 2007). Duncan et al. ( Duncan et al., 2012)
investigated 65 articles on MUVEs, Virtual Worlds
and VLEs and noted six different categories of
published work. These are:
Educational Activities
Learning Theories
Learning Environments
Supporting Technologies
Applicable Age Grouping
Research Areas
Educational activities included virtual field work,
virtual laboratories, collaborative construction,
collaborative simulations, role playing, game based
learning, virtual quests as well as lectures or lessons.
Collaborative Simulation was the most common
with nearly half of the reviewed literature and
collaborative constructional activities were 24%.
Game based learning, virtual quests, role play and
lectures or lessons scored 10% to 20% each in the
reviewed literature. Most articles discussed higher or
further education.
Learning Theory refers to the philosophy behind
the educational approach. According to Twining
(Twining, 2009), experiments have been performed
by integrating different learning theories into
educational activities. Constructivist techniques such
as problem and game based learning are often used
in MUVEs because they allow for experiential
learning. Groupwork scenarios can build on team
strengths and interests and help develop skill and
confidence levels. Games allow for interactivity and
discussion. Collaborative tools such as wikis, blogs,
co-authoring and social networking occur in
MUVEs. Direct instruction, or didacticism, is used
via online lectures for colleges and universities and
webinars, web based seminars, are common in
industry. In terms of student collaboration,
didacticism is at one end of a scale up to knowledge
based constructivism at the most student active end
of the scale. Problem based learning, collaborative,
experimental, instructional, constructivist, didactic
and interactive learning were all identified in the
reviewed literature. Collaborative work was
mentioned in more than 50% of papers, and
knowledge building in 40%. Surprisingly problem
based learning was only mentioned in 14% of
publications.
The Learning Environments category contains
the virtual environments that are used by the
researchers in the reviewed literature. There were
basically two sub-categories; Web 2.0 based VLEs
such as Moodle, Blackboard and WebCT and 3D
web technologies such as Second Life and Active
Worlds (www.activeworlds.com).
The Supporting Technologies category includes
Voice over IP (VOIP), stream video/audio and Chat
and Instant Message (IM). Head-up displays and
immersive technologies can also be included here.
The use of the technologies varies; publishing,
programming or communication are the obvious
functionalities but navigation, mapping and storage
techniques are also included here.
The Applicable Age Grouping demonstrates that
online education is not just for the young. Under 18s
should use Teen Second Life. Experimental projects
and designed activities are different for the separate
age groups; more audio than keyboard support might
be used for the elderly as opposed to the young. Rix
and Twining (Rix and Twining, 2007) define three
levels of pre-18, post-18 and lifelong. Duncan et al.
suggest primary, secondary, further and higher
education followed by lifelong learning and general
education.
The last category includes all the current
research in the field. Researchers are working on
aspects of identity, embodiment and geo-spatial
representation as well as usability, deployment,
knowledge-passing and co-ordination in a learning
environment.
3 REQUIREMENTS FOR
LEARNING SUPPORT TOOLS
In work which investigated the requirements
necessary for a learning to program support tool
(Coull and Duncan, 2011), the result was a set of
ten requirements that programming support tools
should have. Educational tools or mechanisms in
Virtual Worlds will not necessarily be for aiding the
development of coding, but the generic requirements
are undoubtedly similar (see later in 3.1). Students
need guidance and confidence to develop their
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
142
learning online as well as directed, specific support
related to the task on hand. To support a student the
applicable technologies, such as audio, video or the
VLE or access, have to be installed and be reliable.
Further they have to be constantly updated to
maintain up-time when new operating systems or
video cards etc. are installed on servers, or clients,
the network. Further, context relevant physical
support must be available, such as virtual reality
glove sets, joysticks, 3D projections etc. when
students are using virtual archaeological digs or
wandering through reconstructions of ancient
palaces.
In programming terms the ability to write a
correct program depends on four levels of required
knowledge. The first two are known as program
formulation in which the syntactic and semantic
knowledge of a program allows students to develop
a mental model of what is happening in code. The
higher levels of required knowledge are known as
problem formulation; schematic knowledge is the
ability to recognise the patterns or plans to form a
correct solution and strategic knowledge is the
ability to create these plans. Hence a novice moves
from program formulation through to problem
formulation.
It is proposed that a similar framework of
required knowledge is applicable to working in
Virtual Worlds. One level is the requirement that
students understand the context of the task expected
of them – attendance at a lecture, groupwork, a
constructivist activity etc.. Each of these will have
their own expected level of understanding, skill sets
and pre and post conditions for successfully
performing the learning task. A secondary level is
the VW understanding and support requirements
taken from the technology used and understanding
of the computational environment. If a student does
not understand how to navigate through the VW or
store information or search for group members, then
they will be unable to successfully complete the
given task. Also, students must be able to identify
the constructs they require, and perhaps the
communications, to solve their given task and
further, to place these constructs together in an order
which allows a solution to be reached.
Hence, there are four levels of knowledge a
student must have in order to complete a task in a
VW:
Understand the world context and what the
required task is
Understand how to use the functional aspects of
the world, the constructs available to them
Identify and manipulate the constructs they
need to perform the given task
Order these constructs to attain a goal
The required knowledge could be viewed as
basic manipulation or usage knowledge (the first two
bullet points) and schematic and strategic
knowledge, the last two bullet points, where patterns
are recognised and formed into a directed goal.
3.1 Requirements
In the Coull and Duncan paper (Coull and Duncan,
2011) the authors argue for ten emergent
requirements for a support tool for programming
support. These are listed in Table 1, with the
parenthesis in column 2 indicating programming
specific requirements which would not necessarily
be applicable in a VW task.
Table 1: Core requirements of an effective support tool.
Requirement Description
1 (levels of support)
Present (both standard compiler
and enhanced) support concurrently
2 (linkage)
Link to teaching resources as a
means of information delivery and
student-tutor dialogue
3 (errors)
Identify and advise on commonly
observed (semantic) errors
4 (key constructs)
Embody knowledge of key
constructs needed to solve a
problem
5 (construct relationship)
Embody knowledge of the
relationships between the
constructs needed to solve a
problem
6 (dissemination strategy)
Knowledge should be disseminated
to students in successive stages
7 (variant solutions)
Ensure that the knowledge
accommodates variant solutions, if
they exist
8 (varied problems)
Provide support for different
problems
9 (Support reduction)
Support may be progressively
reduced over the teaching period
10 (Volition)
Use of the tool must be voluntary
on the part of the student
The descriptions of the generic support tools
indicate that the tool must be voluntary (10), be
provided concurrently with the normal system
support (1) and be progressively reduced throughout
the teaching period (9) so that students do not
become reliant on it. There should be links to
teaching resources such as lectures, laboratory
exercises, tutorial sheets and perhaps a chat or email
mechanism to the course tutors (2). Knowledge of
the key constructs in the system, or world, must be
given and practiced (4) and the relationships
between those constructs must be understood during
the experience (5). This knowledge may be
disseminated via a staged process (6). As the course
or module is developed, different tasks should be
REQUIREMENTSFOREDUCATIONALSUPPORTTOOLSINVIRTUALWORLDS
143
embodied within the VW so that it is not a one-off
experience (8) and when marking or grading student
work, tutors, or an automated grader, should take
into account that solutions may be varied (7). On the
functional support side, the system or at worst, an
email account, should be available for logging faults,
either with the computational aspects, the physical
support, the tasks given or semantic problems to do
with the rendering etc. (3). These can therefore be
dealt with by the tutors and support staff.
The requirements for learning support are
therefore divided between tool support, task support
and learning support. These could then be codified
as:
Tool support – Present tool alongside original
functional support, make the use voluntary,
reduce the support throughout the course
lifetime and collect (and fix) errors.
Task support – Knowledge of constructs and the
relationships between them, if any, must be
given.
Learning support – Links to teaching resources
should be given and knowledge should be
disseminated at appropriate intervals, varied
tasks should be attempted and grading should
take into account the varied successful
outcomes.
Each of these are necessary to allow students to
learn virtually and to progress through given tasks in
a virtual world. However a task is envisaged or
developed it has to be graded and as with any
qualitative analysis in a computational environment,
this has to be tackled via a variety of evaluation
procedures.
4 EVALUATION ISSUES
Evaluation is a difficult task and it can often take a
significant number of hours to do when the product
is not as straightforward to analyse as a functional
code procedure. As with anything visual, a measure
of qualitative analysis has to be used alongside
quantitative analysis. Steinkuehler and Duncan
(Steinkuehler and Duncan, 2008) state that some
user posts on a World of Warcraft forum
(www.worldofwarcraft.com
) displayed an evaluative
position where “knowledge is an open-ended
process of evaluation”. Some users demonstrated
an absolutist approach treating knowledge as
“objective, certain and .. accumulative” whereas
others were relativistic and treated the world as
subjective as in “to each his own”. Therefore it can
be concluded that this type of qualitative analysis
has to be garnered from questionnaires, interviews
and posts and is difficult to analyse because of its
divers nature.
Gabbard et al. (Gabbard et al., 1999) describe a
methodology for evaluating virtual environments
using user task analysis, expert guidelines-based
evaluation, formative user-centered evaluation, and
summative comparative evaluations. They discussed
cycles of evaluation with an expert on hand to
determine guide-lines for usage. Most of this work
considered usability characteristics and attempted to
determine summative scores for formats or
placements of information. de Feitas et al. (de freitas
et al., 2009) advocate incremental testing alongside
evaluation of virtual world learning experiences.
Their work used an inductive methodology to
construct theories and explanations based upon
observations and survey data. They monitored
students and used surveys, specifically using chat
logs and video footage. They set out to test their
Four Dimensional Framework which includes a
learner specific evaluation (profile, role,
competency), the pedagogy (associative, cognitive
and social/ situative), the representation (fidelity of
design to implementation, interactivity, immersion)
and the context (environment, access to learning and
supporting resources).
Consequently, evaluation of virtual worlds for
learning require a multi-faceted approach wherein
usability guidelines have to be stated and then
measured for fidelity as well as for content,
suitability, affectiveness, technical and pedagogical
appropriateness. Functionality of the required and
student developed solution have to be compared and
further, the physical support side has to be
considered in terms of the network speed, server
connection and up-time and standards etc..
Therefore a levelled approach is advocated as in
section 3.1, but including the network layer:
Network support – Network statistics must be
gathered and compared to a standard or
expected level of quality.
Tool support – The underlying tool and/or
virtual world must be evaluated to determine if
the necessary learning constructs, the required
learner support and ability to add learning or
virtual constructs exists.
Task support – The virtual constructs to
engender a learning experience must exist or be
able to be built with given or constructed
learning.
Learning support – The required knowledge
must be linked to or in evidence.
Further work on evaluating specific learning
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
144
experiences in Second Life and other virtual worlds
is necessary to create a taxonomy of issues and
determine if a generic support framework can be
determined from the above. Effectively the
computational side, the network and the virtual
world support, must be evaluated by quantitative
analysis based on quality of service and usability.
The learning task requires qualitative analysis based
on pedagogy and pre and post knowledge derived
from questionnaires, interviews and logging
information.
The core requirements mentioned above are
currently undergoing evaluation in a variety of
virtual world learning scenarios and systems and the
results will be recorded in due course.
6 CONCLUSIONS
Virtual Worlds are increasingly being used for
teaching with constructivist and collaborative
learning being the common approaches taken.
Consequently it is imperative to develop
requirements check-lists and evaluation mechanisms
for the learning aspects as well as the virtual world,
any support tools and the underlying system
requirements.
The paper has outlined two papers on the
educational uses of virtual worlds and the generic
requirements for a support tool. In marrying the
support requirements to the uses for virtual worlds a
layered approach to determining requirements and
evaluating virtual worlds for educational purposes
emerges in which both qualitative and quantitative
analysis is required at network, system, virtual
world, learning task and user levels.
REFERENCES
Akpan, J. I. & Brooks, R. J. Year. Experimental
Investigation of the Impacts of Virtual Reality on
Discrete-event simulation. In: Kuhl, M. E., Steiger, N.
M., Armstrong, F. B. & Joines, J. A., eds. Winter
Simulation, 2005 Orlando, Florida. ACM, 1968-1975.
Childs M. 2007. Real Learning in Virtual Worlds.
Warwick Interactions Journal, 30, 38-45.
Coull N. J. & Duncan, I. 2011. Emergent Requirements
for Supporting Introductory Programming. ITALICS -
Innovation in Teaching and Learning in Information
and Computer Sciences, HEA e-journal.
Dawley, L. 2009. Social Network Knowledge
Construction. On the Horizon, 17, 109-121.
De Freitas, S., Roebolledo-Mendex, G., Liarokopakis, F.,
Magoulas, G. & Poulovassilis, A. 2009. Learning as
Immersive Experiences: Using the Four-Dimensional
framework for Designing and Evaluating Immersive
Learning Experiences in a Virtual World. British
Journal of Educational Technology, 41, 69-85.
Duncan, I., Miller, A. & Jiang, S. 2012. A taxonomy of
virtual worlds usage in education. British Journal of
Educational Technology, to appear 2012.
Edirisingha, P., Nie, M., Pluciennik, M. & Young, R.
2009. Socialisation for learning at a distance in a 3-D
multi-user virtual environment. British Journal of
Educational Technology, 40, 458-479.
Gabbard, J. L., Hix, D. & Swan II, J. E. 1999. User-
Centered Design and Evaluation of Virtual
Environments. IEEE Computer Graphics and
Applications, 19.
Getchell, K., Miller, A., Allison C., Kerbey, C., Hardy R.,
Sweetman, R., Crook, V. & Complin, J. 2006. The
LAVA Project: A Service Based Approach to
Supporting Exploratory Learning. IADIS International
Conference WWW/Internet Murcia, Spain
Jenkins, M., Browne, T. & Walker, R. 2005. VLE surveys:
A longitudinal perspective between March 2001,
March 2003 and March 2005 for higher education in
the United Kingdom. UCISA.
Mason, H. Year. Experimental Education in Second Life.
In: Livingston, D. & Kemp, J., eds. SLCC' 07, 24-26
Aug. 2007 2007 Chicago. 14-18.
Oliver, M. & Carr, D. 2009. Learning in virtual worlds:
Using communities of practice to explain how people
learn from play. British Journal of Educational
Technology, 40, 444-457.
Rix J. & Twining P. 2007. Exploring education systems:
towards a typology for future learning? Educational
Research, 49, 329-341.
Salmon G. 2009. The future for (second) life and learning.
British Journal of Educational Technology, 40, 526-
538.
Steinkuhler C. & Duncan, S. 2008. Scientific Habits of
Mind in Virtual Worlds. Science Education
Technology, 17.
Twining, P. 2009. Exploring the Educational potential of
Virtual Worlds - Some Reflections From the SPP.
British Journal of Educational Technology, 40, 496-
514.
REQUIREMENTSFOREDUCATIONALSUPPORTTOOLSINVIRTUALWORLDS
145
