Repurposing social networking technologies to

encourage preservice teacher collaboration in

online communities: A mixed methods study
Michael Moroney
Universiti Brunei Darussalam

The purpose of this study was to determine the extent to which

trainee teachers were able to use social networking technology for
online collaboration in a learning community. Communities of
practice are at the forefront of inservice teacher professional
development and their usage in preservice courses may have value
also, if for no other reason than to provide opportunities for
participants to learn how to collaborate effectively online. This
study used a mixed method approach to determine the extent to
which participants were totally committed to online collaborative
learning. Findings indicate that participants generally supported
the concept of an online community but that they found it difficult
to develop and sustain one. There were also indications that the
perceived value of online collaborative learning was in many cases
not able to overcome long held practices and beliefs that value
competition and individual effort. Thus the concept of online
communities of practice may more easily be introduced to inservice
teachers who already have working experience of collaboration and
cooperation upon which they can draw or when the high stakes
examination system, in which participates operated, is replaced by
one that measures the transaction of learning. Directions for
further research are discussed.

Introduction

There is considerable interest in collaborative learning, both amongst

teachers and amongst students in classrooms. There is support for this
model if for no other reason than because a team working together may
attain greater overall achievement than that achieved through
individualized learning (Garrison & Anderson, 2005; Soller & Lesgold,
2007). Garrison et al. (2005) stated “it is a practical necessity” that
students be independent thinkers, interdependent collaborative learners,
and that these characteristics are “…core values in any educational
experience” (p. 22). Carugati et al. (2008) further highlighted the
importance of collaborative learning when they described web groups and
social networks as two key concepts that best depict online e-learning. In
a web group environment participants use an Internet site as a repository
for shared resources. Social networks, on the other hand, may be
considered Web 2.0 technologies that afford participants the opportunity
to participate in joint knowledge creation through a process of negotiation
with members of an online community. Web groups are often associated
with “Web 1.0” technologies and social networks are often associated with
Web 2.0 technologies. Online facilities such as Facebook, MySpace,
Secondlife and Ning are examples of Web 2.0 technologies.

1
 It is recognised that teachers benefit from collaborative learning just
as students do. For example, those teachers engaged in professional
development learn best when they are able to work in teams on common
problems over extended periods of time (Loucks-Horsley, Love, Stiles,
Mundry, & Hewson, 2003; Loucks-Horsley & Matsumoto, 1999; Timperley,
Wilson, Barrer, & Fung, 2007). Technology, in this situation, can be used
to add value to the quality of the professional development experience by
facilitating more and different opportunities to share ideas and to enhance
collaboration. In some education systems, such as the secondary school
system in Brunei, it is often the case that a school may only have one or
two specialist teachers. For example, in Brunei it is uncommon for any
secondary school to have more than one chemistry teacher. Furthermore,
schools are geographically far enough apart that it is not feasible for
teachers to gather together regular professional development sessions.
Therefore it is of benefit to teachers if online team collaboration can take
place to supplement infrequent face-to-face meetings and improve the
effectiveness and relevance of teacher professional development.

However, this said, online CSCL environments are complex places

and online communities may not always meet the expectations of their
members (Salmon, 2005). Therefore there is the need to identify the
factors that do help to ensure online communities are successful.

Community members of any online learning group may be broadly

categorized as being: non-compliant, procedurally compliant or totally
committed to online collaborative learning. For example a non-compliant
team member may choose not to participate in community activities in
any meaningful way. Similarly, a procedurally compliant member is also
not especially useful as he will only do what is required to maintain the
appearance of being a good member, while actually contributing very
little. For example a procedurally compliant student may contribute the
required number of postings to an online forum to avoid being in default of
course requirements whilst at the same time making little effort to provide
content that advances the collective knowledge of the community. The
best environment for all concerned is one in which all members are totally
committed to CSCL, for that is when a group will have the best chance of
meeting the expectations of its members. Therefore it is worth the effort
to find out what the factors are that will maximize community
performance.

Bearing in mind the above qualifying comments it is arguably

important therefore to explore CSCL communities from two related
perspectives; by seeking to understand how CSCL can be implemented
and by exploring the means by which we might tell that collaborative
learning is taking place.

Research into computer supported collaborative learning (CSCL) is

often explored from the theoretical perspective of social constructivism.
This epistemological position favours the use of discourse by members of
a team to construct individual meaning through the construction of shared
artifacts, for example through the development of online forum
discussions and wikis (Dougimas & Taylor, 2002). From the social
constructivist perspective a group of learners will naturally gravitate

2
towards other like-minded colleagues to form a community of practice
(COP) (Wenger, 1998; Wenger, McDermott, & Snyder, 2002). For
example, a group of chemistry teachers may perceive the need to create
new and different learning opportunities for students that have difficulty
grasping complex concepts. By working together the teachers are more
likely to mitigate the complexity of creating technology rich lessons for
their students than if they worked alone (American Association of Colleges
for Teacher Education & Committee on Innovation and Technology, 2008).

Newcomers to a community, such as the chemistry teachers

community described in the example above, go through a learning and
acceptance process in which they must master the sociocultural practices
of the COP they are joining (Lave & Wenger, 1991). Furthermore Lave et
al. (1991) reported that it is only after this “apprenticeship” period that
newcomers can begin to reap the benefits of belonging to a COP. Lave &
Wenger describe the apprenticeship process newcomers go through as
development of legitimate peripheral practice (Lave & Wenger, 1991).
Wenger et al. (2002) reported that it is important to adopt practical
models for COP development if communities of practice are to reach their
full potential, it is not enough to leave development of a community to
chance.

…communities typically undergo several changes in their focus,

relationships and practice. They commonly shift from sharing ideas
and tips to stewarding their practice – building refining, and
expanding the domain and its relationship to other domains. They
move from a lose network of personal relationships to a group with
a common sense of identity, combining intimate knowledge of each
others’ approach with a sense of collective responsibility for the
domain.
(Wenger et al., 2002, p. 111)

Several researchers have developed checklists that can be used to

guide the development of a COP. For example, Kopp & Mandl (2008)
reported four common characteristics found in effective (CSCL-COP)
groups: (1) goal orientation e.g. “I indentified myself with the group goal”,
(2) task completion “the priority was task solving”, (3) cohesion e.g. “we
communicate freely”, and (4) taking responsibility e.g. “I tried to complete
assigned tasks”.

Additionally, Soller & Lesgold (2007) described four strategies for

effective group collaboration: (1) construction, (2) criticism, (3)
accumulation, and (4) motivation. From the construction perspective
community participants make their own new knowledge by observing
group action (that imparts new knowledge). From the criticism
perspective, individuals help each other overcome the discomfort of
cognitive dissonance through exploring the reasons behind alternate
beliefs and attitudes. From the accumulation perspective, a broad
knowledge base can be quickly developed and sustained when group
members contribute concurrently. And fourthly, from the motivation
perspective students engage in social comparison in which they seek to
measure their own actions against the actions of others (Soller & Lesgold,
2007).

3
 The checklists just described are useful for developing communities
of practice and do in deed complement each other. However, these 8
learning environment characteristics are insufficient to guarantee
successful online collaboration. For example, Salmon (2005) suggests that
any successful online collaborative learning community will likely have a
skilled e-moderator to manage it, the implication for preservice teacher
professional development being that it is sensible to ensure that groups
are able to acquire or develop a skilled moderator.

Aim of this research

The particular study outlined below in this paper is part of ongoing

research into CSCL being conducted with preservice and inservice
teachers in Brunei Darussalam. Much of this research relates to the use of
computer supported collaborative learning environments for the
professional development of inservice and preservice secondary school
teachers. There are many questions that the researcher is trying to
answer but the major question addressed in this paper is whether or not
preservice teachers totally commit to computer supported collaborative
learning or whether the context in which their learning takes place inhibits
participation beyond “procedural compliance”.

Significance of the research

This research should add to an understanding of how preservice teachers
can be encouraged to adopt CSCL by identifying the factors that either
afforded or constrained CSCL community development in this particular
setting.

Method

The paper reflects a pragmatic worldview that focuses on practical

outcomes that might lead to change in practice (Creswell & Plano Clark,
2007). At the same time it acknowledges the importance and utility of the
constructivist, advocacy/participatory and postpositivism worldviews
described by Creswell and others (Creswell & Plano Clark, 2007;
Tashakkori & Teddlie, 2009).

To structure this research a modified parallel mixed model design

was chosen as the best fit (Bryman, 2008; Creswell & Plano Clark, 2007;
Tashakkori & Teddlie, 2009). This model requires qualitative methods
and quantitative methods to be collated and discussed separately in
relation to findings before being drawn together in an overall meta-
inference (Tashakkori & Teddlie, 2009). As shown in Figure 1, overall
results of the study were informed by the separate quantitative and
qualitative strand results. The qualitative strand involved analysis of forum
and weblog transcripts. The quantitative strand involved analysis of
survey results and web server records as well as quantitative analysis of
Ning and wiki sites.

4
Figure 1: Modified parallel mixed methods design

Participants

68 participants contributed to the study. All were enrolled in the

same educational technology course that ran for 14 weeks at the
University of Brunei (UBD) in 2009. 18 (26%) participants were male and
50 (74%) were female. 61 (89.7%) students were in their fourth year of
study and 7 (10.3%) were in their third year. Students came from
different disciplines including 10 (14.7%) majoring in Biology, 10 (14.7%)
in Chemistry, 13 (19.1%) in Economics, 11 (16.2%) in English Literature,
10 (14.7%) in Geography, 8 (11.8%) in History, 5 (7.4%) in Mathematics,
and 1 (1.5%) in Physics. All students were undergraduates enrolled in
either B.Sc. Education or B.A. Education programmes.

Participant CSCL activities

As a part of their coursework participants in the study were required

to work in teams to build and manage an online community of practice
presence. For example, a group of 10 preservice chemistry teachers
formed a team to build a site that was used by “O’ and ‘A’ level chemistry
teachers.

Participants formed 19 small groups responsible for repurposing an

online social networking site called a Ning. Additionally, groups decided
they needed a way to organise online content that was easily navigable
and accessible to online visitors, and for this purpose most groups chose
to link their Ning to a wiki. The combination of a Ning and a wiki allowed
groups to manage the social collaborative aspects of their community on
the Ning site while the wiki site was used to manage content.

Data Collection

5
 Qualitative and quantitative data was collected as a part of a parallel
mixed methods design. Several data sources were used:

• Participant weblogs and online forum transcript files were used

as data for the study.

• A survey tool called the Web-based Learning Environment

Instrument (WEBLEI) was administered to gather quantitative
data about student perceptions of their CSCL environment.

• Web server logs were used to gather data about both

individual and group online practice.

Qualitative data collection

Online forum contributions

Participants were asked to contribute to online forum discussions

about their CSCL experiences by addressing specific questions posed on
their LMS site.

Weblog contributions

Students were required to maintain an online journal in which they

were to write down their thoughts about the affordances and constraints
of their CSCL environment.

Participant weblogs and online forum transcripts were collected from

a web server and directly imported into NVIVO 8 for content analysis.

Quantitative data collection

WEBLEI

The WEBLEI is a 31 Likert item survey instrument that has four

subscales (see Table 1). It was used to collect data about student
perceptions of their CSCL learning environment. The instrument was
administered online using a survey tool called LimeSurvey (see
http://www.limesurvey.com). Data was automatically collected in
electronic form and exported into an SPSS 15 data file for statistical
analysis.
Table 1
WEBLEI scales

Scale Scale meaning and example

Access scale Measures the extent to which students believe they can use
technology as and when they want to.
I can access the learning environment at times convenient to
me.

Interaction scale Measures the extent to which students believe that the CSCL
environment facilitates collaborative learning
Other students respond promptly to my queries.

6
Response scale Measures the extent to which students experience a sense of
achievement and satisfaction in the CSCL environment.
I enjoy participating in discussions in this learning
environment.

Results scale Measures the extent to which students believe the CSCL
environment is structured to facilitate achievement of course
learning objectives.
The expectations of tasks and roles are clearly stated in the
online environment.

Web analytics data

Web analytics is the examination of web server data logs. Web logs provided a variety
of information about participant online behavior including the types of activities participants
engaged in as well as data about the frequency and timing of online interaction. In this study
it was possible to use very detailed LMS (Moodle) logs and less detailed, but never-the-less
useful Ning data. Ning data included information about the number of and the timing of
online discussion contributions as well as details about changes to site content (such as the
addition of videos, new blog postings and online polls).

Ethical considerations

Much data for the study was available in the public domain. That is
to say anyone could enter the correct URL into their browser and go and
explore the data generated by study participants.

There is an argument that as long as participants have been

informed that they make contributions to a public space when they go
online then there is less obligation on the part of researchers to “protect
the anonymity of individuals using the venue, or to seek their informed
consent” (Bryman, 2008, p. 654). Easily accessible electronic data is
generated, whether planned for or not, whenever anyone goes online.
The ethical question a researcher faces is therefore not whether data
should be collected, but what to do with data that will inevitably end up in
the logs files of an Internet web server.

Data Analysis

Analysis of Web logs and online discussion transcripts

Web logs and online discussion contributions were analysed qualitatively

with NVivo 8, following the procedures outlined in Lincoln and Guba (1985)
and Glaser and Strauss (1967). That is to say participant responses were
separated into meaningful units where a unit was an idea, observation or
thought. Responses were organised under theme headings. Themes
were grouped into subcategories in an iterative and cyclical process
during and after initial and repeated reading of transcripts. Participant
contributions were gradually reorganized into a logical thematic taxonomy
that could be collated and reported.

7
Analysis of the WEBLEI

Internal consistency and discriminant validity were calculated. Averages

for Likert scale responses were calculated for WEBLEI subscales and these
formed the basis for further discussion and comparison with the results of
other research strands.

Web analytics

Analysis of web server logs revealed participant usage patterns, including

data about the number of postings individuals contributed to online
discussions, whether to a group Ning or to an LMS discussion thread.
Analysis of Nings revealed the levels of participation of COP members and
the number and type of resources placed online. Analysis of logs also
revealed the ratio of discussion thread postings to views.

Limitations

The study used a convenience sample and therefore results cannot

reasonably be generalised to other populations. Furthermore, the study
is a snapshot of participant actions in week 10 of a 14-week semester.
Therefore the study cannot be used to report, for example, whether or not
the quality of collaboration in groups was improving over time. The
WEBLEI is a learning environment that was administered online for the
first time. The implications of conducting the survey online indicate that
the WEBLEI might be improved. Only one researcher carried out content
analysis and therefore the findings were subjectively interpreted.

Findings and Discussion

In this section the findings for each research strand are reported
and inferences made. Concluding remarks bring together the
individual findings for each strand in a meta-inference section.
Forums and Blogs

Content analysis revealed that participants identified 53 factors that

impact on the development of online communities of practice. These
factors were collapsed into three major categories: (1) reasons to connect,
(2) enablers, and (3) limiting factors. A reason to connect is necessary if
widespread adoption of CSCL is to take place and an online COP is to be
sustained. That is to say an online site needs to contain relevant content
that will help participants to solve the educational problems that
participants visit the site to find solutions to. Secondly, enablers were
identified that could facilitate participation in an online community of
practice. Thirdly, participants identified limiting factors that they felt
prevented (or at least inhibited) active online learning in a community of
practice.

Reasons to connect

Table 2 lists the main reasons why participants wanted to connect to

their online social network. Participants found that there were many
reasons for wanting to go online but that the most powerful motivation
was to exchange ideas with other group members. They exchanged ideas
for two reasons: (1) to learn new things, and (2) to confirm what they

8
already knew. Participants also reported that they valued active learning
rather than being passive receivers of content and that they understood
that the actual process of contributing to online discussions was
educationally valuable.

As the following quotation shows, participants often enjoyed

engaging in online discussions, even when they were exposed to ideas
and viewpoints that challenged their own understanding.

Participant 5 “-online community makes exchanging

ideas a breeze (huge collection of tasty
ideas… yummy)”

However, despite claims of enjoying CSCL content analysis revealed

that much of the time the online COP facility was used as a convenient
meeting place where operational strategies unrelated to CSCL were
discussed. For example, a discussion by an Economics group extended to
52 replies as they grappled with the operational aspects of getting a group
task completed on time:

Participant 1: lets just meet on saturday!!! =)

Participant 2: Okay, we have decided yesterday on what task we are

going to give for each one of us... so I just updated it
here
For Compiling & Powerpoint: Khatib, Farhan & Noh
Barter Trade: Qieqie & Iffah
Money: Zai, Rina & Qilah
Classroom Management & Assessment: Liyana & Salina”

This particular discussion thread quotation highlights another

interesting factor – that measuring the quality of discussions by counting
the length of posts or the number of posts in a discussion thread may not
always be an accurate way to measure the effectiveness of group
collaboration.
Table 2: Reasons to connect

Motivation
1. Exchange of ideas

2. Active learning

3. Challenging ideas

4. Knowledge expression

5. Clarification and confirmation

6. Control over pace of learning

7. Critical thinking and related higher order

thinking skills

8. Interpersonal skills development

9
 9. Reflection

10. Convenience of online communication

Participants not only enjoyed listening to what others had to say,

they also liked to use the forum to confirm their own understanding.
When other group members questioned the posting then it may have
indicated that the understanding behind the original posting needed
further development. Further individual reflection might take place, and
as understanding evolved then this was reflected in the transcripts…
before a revised version of the original posting is published

Enablers

Participants described 6 essential community of practice enabling factors:

(1) desirable membership characteristics, (2) high quality online content,
(3) members that participate regularly, (4) a code of practice (explicit or
implicit), (5) members can participate easily, and (6) members play an
active part in supporting the community (see Table 3). Open mindedness
was rated as the most important characteristic for members to have in an
online community. The remainder of this section summarises the views
participants held about COP enablers.
Table 3: Enablers of successful CSCL networks

Enablers
1. Member characteristics

o Open minded

o Members are a part of a community

o Creative contributors

o Well mannered

2. Content quality

o Fresh attractive and interesting

content

o Use of different types of media

o High quality relevant content

3. Participation level

o Active participation is required

o Continuity of participation is
necessary

o Forced participation may be a good

thing

4. Code of practice

10
 o Members are encouraged to contribute

o Members have equal rights

o Members stay on the topic

o Different viewpoints are respected

o Members may contribute anonymously

5. Connectivity

o Can connect anytime and from

anywhere

o Bandwidth is appropriate

o The technology is easy to use

6. Member responsibilities

o All have a responsibility to support

the community

o Participants promote new

membership and engagement

o Old hands help newcomers

Team members felt that they needed their online community to be

moderated by an administrator, and that one of the administrator’s main
jobs was to coordinate quality control activities. The moderator was
therefore “powerful” but also under constant “surveillance” by group
members to ensure they were doing their job properly. Participants
mentioned that the needed a commitment to continual improvement and
a willingness to change content when understanding of the affordances of
technology changed. For example, as newcomers to a COP became more
knowledgeable about the technologies they were learning to use they
made quite radical changes to the appearance and navigation structures
of their group wiki and Ning. This process of going through content
revisions appears to be a natural event, and might be a useful predictor of
how developed an online COP has become. In the start groups were keen
to upload as much content as possible and place it online without much
thought for accessibility of quality control. As the collective knowledge of
their COP developed they learned how to control quality better and to
ensure that their site design was user friendly and focused towards the
needs of a particular audience.

All groups decided that they would like to upload material into a
“knowledge base”. They felt that there needed to be a collection of online
resources available for COP members and visitors to download. Eventually,
all groups converged towards a common solution to their COP and CSCL
needs, a solution that put the social aspects of their community with the
Ning and a knowledge base on a separated wiki site.

Limiting factors

11
 This section describes the perceived barriers COP development
identified by participants.

Although all participants felt that it was worthwhile to participate

online there were barriers that prevented full member participation. Five
general themes related to barriers emerged from the data: (1) technical
difficulties, (2) learning preferences, (3) social network dysfunction, (4)
participant competencies, and (5) member perceptions. Table 4 lists the
major concerns participants mentioned in their blogs and online forum
discussion threads.

Table 4: Factors that prevented full membership in a COP

Limiting factors
1. Technical difficulties

o Limited Internet access

o Insufficient member ICT skills

2. Social network dysfunction

o Members did not know social network etiquette

o The aims of the social network were too broad and

general

o Social network administration was ineffective

o Poor content quality control

3. Member perceptions

o Unequal participation levels

o Cultural conformance

o Lack of interest

o Mistrust

o Procedural conformance

o Irregular engagement

4. Participant competencies

o Undeveloped elaboration skills

o Lack of ICT skills and knowledge

5. Learning preferences

o Prefer face to face (f2f)

o Prefer to learn individually

Technical difficulties

12
 Somewhat surprisingly, given the perceived widespread availability
of Internet access in Brunei, 12% of participants reported technical
difficulties that prevented adequate access to online facilities. They
reported that they either did not have ready access to the Internet outside
University or that quality of Internet bandwidth was unfit for purpose. For
example, several students mentioned that it was good to be able to view
educational videos and animations, but that it was an unreliable activity
when bandwidth could not be guaranteed.

Social network dysfunction

Student familiarity with social networking tools such as Facebook

may have hindered attempts by group members to develop group
cohesion and to define community purpose. Some students talked about
things that had nothing to do with the task at hand, and seemed to be
unaware of group focus, purpose, or the idea of a COP. In some instances
it was as if the COP was more or less an extension of Facebook with
students making use of the extensive range of design options for the sake
of design rather than to solve an educational problem. Organizational
issues, especially poor moderator performance, also prevented realization
of COP potential. It is, for example, important that moderators ensure that
their online facility is easily navigable and clearly focused. It may well be
that social networks are reasonably easy to set up and moderate.
However, it is as well to remember that the nature of social interaction
needs to change when the technology is repurposed for use in a COP, and
for this to happen the moderator needs to show considerable knowledge,
skill and leadership.

Member perceptions

Participants raised a number of concerns that perhaps relate directly

to the higher education learning environment rather than to the
affordances and constraints of online collaborative per se. Participants
resented the fact that some group members did not contribute,
particularly when members had a stake in seeing that their COP was
successful. One participant was eventually ejected from his group (by his
group) because he didn’t contribute meaningfully online. Two online
discussions highlighted a lack of trust as being behind reticence to
contribute. Participants did not want to “risk” publishing their thoughts.

Interestingly, one group reported that cultural conformance was a

stumbling block that hindered COP development. Brunei participants were
self-described as non-confrontational, empathetic and polite to a fault. It
was felt that participants were so well mannered that they would not
contribute a view that was contrary to a view already published. A
characteristic such as this is clearly going to limit development of an
online COP.

Participant competencies

Several knowledgeable participants reported being somewhat fed-up

with having their time wasted by a minority of people in large group
forums that didn’t make meaningful contributions to discussions.
Newcomers to discussions often repeated what had already been said,
contributed information that was off the topic or otherwise didn’t
contribute to the collective knowledge of participants. One participant

13
correctly suggested that this problem was because inexperienced
contributors didn’t know better and weren’t being shown how to change.

Some participants reported that they were not confident about their
elaboration skills. Although all participants had proven English language
proficiency none could claim English as their home language. As one
would expect, English literature majors were considerably more fluent
defending a position online than participants training to be science,
mathematics or geography teachers. Clearly this situation will always
have a negative impact on a COP, as it will reduce the variety views and
the quality of discussions upon which CSCL is based and which generates
the collective knowledgebase at the heart of a community. Therefore
overcoming this barrier should be a high priority within a COP community
and every effort should be made to identify and support COP members
that may feel this way.

Learning preferences

An unexpected finding was that some participants simply didn’t like

computers and didn’t want to use them, and given the opportunity they
would prefer to work face to face to complete project work. This reticence
to contribute impacts on the individual concerned and also the other
members of his team and so it will be worthwhile to find ways to change
the perception of preservice teachers that resist the use of technology.

Another learning preference perception revealed in forum discussions

was that some participants preferred to work individually and not in a
group. Such participants were resistant to CSCL (or collaboration of any
kind) and therefore may be more difficult to convince that CSCL is
worthwhile. There are two concerns here: (1) these participants will be
teaching in schools in an environment in which they are expected to
support collaborative learning pedagogy (something they do not believe
in), and (2) these participants may prefer not to contribute meaningfully to
professional development teams. These are important issues to address,
the consequences of impact negatively on the future learning of students
and staff in schools (professional practice and professional development)
as well as current preservice teachers.

Quantitative findings and discussion

WEBLEI

Table 5 summarises the findings of statistical analysis of WEBELI

data.
Table 5
WEBLEI subscale descriptive statistics and reliability scores

Mean
Numb Alpha correlation
er of Reliabilit Scale Scale with other
Subscale Items y mean SD N scales
Access 7 0.70 3.77 0.28 63 0.35

Interactio 0.22
n 8 0.66 3.08 0.32 64

14
Response 8 0.64 3.71 0.24 65 0.34

Results 8 0.86 3.82 0.19 65 0.34

The WEBLEI was administered online with the participation of 65

(95.6%) respondents. Internal consistency (alpha reliability) was obtained
from the sample as indices of scale reliability. The alpha coefficient scores
ranged from 0.64 to 0.86 in this study as compared to 0.68 to 0.87 in a
similar type of study carried out with the same instrument in Singapore
(Chang & Fisher, 2003b). On the whole the values for this study were
acceptable and comparable to the previous study.

The mean correlation of each individual scale with the results of the
other three scales was used for discriminant validity. WEBLEI discriminant
values ranged from 0.22 to 0.35 in this study, as compared to 0.37 to 0.49
in Chang and Fraser (2003a). On the whole the values for this study were
comparable to the previous study.

The first sub-scale, access, measured participant perceptions of how

easily they could physically connect to their social network site and the
extent to which they could make their own decisions about the extent to
which they participated in online activities. The average participant scale
value for access was 3.77, which is near to “often” on the WEBLEI Likert
scale. This finding confirmed that participants were mostly able to
connect as and when they wanted to. This was an important finding as
the access scale result may well have a bearing on the remaining three
scales in the sense that if access responses were poor then it is likely that
interaction, response and results scale responses would be poor also.

The second subscale, interaction, measured participant perceptions

of the quality of online discussions. The average participant scale value
was 3.08, which is close to “sometimes” and noticeably lower than the
average for the access subscale. This was an interesting result because it
may mean that participants were able to get online easily, but once there,
didn’t know how to collaborate effectively. This result further highlights
the need for preservice teacher educators to be proactive and systematic
about developing and integrating online collaborative learning
environments. Although participants may have had considerable social
networking experience it appears that this alone was insufficient for them
to be able to successfully repurpose a social networking facility (Ning) for
online collaboration. Wenger et al. (2002) describe the challenges of
facilitating threaded discussions:

…when a Web-based conversation goes through natural

cycles of low activity, it may look like the community is dead.
When members think the community is waning, it starts a
vicious cycle. Who wants to contribute if no one is listening?
The community coordinator can stimulate these discussions
by occasionally stirring the pot with an intriguing question or
a provocative statement. Such interventions can return the
community to the awareness of members.

(Wenger et al., 2002, p. 131)

15
 The third subscale, response, measured participant perceptions of
their overall level of satisfaction with their online CSCL experience. The
average participant scale value was 3.71, which is close to “often” on the
WEBLEI Likert scale. This indicated that students mostly enjoyed working
in the online environment. However, analysis of individual items in this
subscale revealed that most participants reported enjoying participation in
online discussions only “sometimes”. Furthermore, they also reported
that they only enjoyed working in groups “sometimes as well. There are
two item questions that arise out of these two particular item responses:
(1) if students were taught how to collaborate online effectively would this
increase their enjoyment when they work online, and (2) given that all
participants consider a language other than English as their home
language would the enjoyment result have been different if students had
been able to write in the mixture of Malay and English they prefer to use
when conversing online at social networking sites like Facebook.

The last sub-scale, results, measured participant perceptions of

whether or not the CSCL environment was structured in a way that would
help them to achieve personal learning goals. The average participant
scale value was 3.82, which is reasonably close to “often” on the WEBLEI
Likert scale, an unexpected finding given the lower scores for items that
related to satisfaction with the quality of online discourse in forums.
Furthermore, the result seems to contradict the web analysis findings that
report relatively low group forum participation, an indication of only
limited COP cohesion.

Web analytics

Analysis of forum activities revealed that the 68 participants viewed

discussion postings a total of 8,812 (92.69%) times and contributed to
online forums 695 (7.31%) times. On average, each participant made
10.2 individual postings to discussion threads and viewed 130.0 online
forum discussion threads. The most active online participant contributed
61 postings and viewed 445 contributions during the study. In this case
the participants’ percentage ratio of postings to views was 13.12%.
However, at the other extreme there were 4 participants that made no
contribution to either a Ning or LMS forum (although the average number
of views for these four participants was 47.5).

The results from paired sample t-test analysis indicate that there is a
significant difference between the number of contributions made to team
discussions and the number of contributions made to whole group
discussions t(67)=4.17. p<0.05 with the mean total number of postings
for small teams and all 68 participants together being 4.06 (SD=4.22) and
6.16 (SD=5.73) respectively. This results indicates that participants
contributed significantly less often to their groups than they did to the
collective forum. These finding somewhat supported the notion that many
participants did not fully commit to CSCL and that there was even less
commitment in the small group communities that did not have the same
level of community moderation as present in the LMS general forum. This
does not mean that participants think CSCL is a bad idea; in deed content
analysis findings and WEBLEI findings support the view that participants
have a high regard for CSCL. What was apparent however was that
participants did not need the “computer supported” element of CSCL
because they had ample opportunity for face-to-face team collaboration.

16
Therefore the COP side of participant websites remained largely
undeveloped while at the same time the online content (wiki) aspect of
group sites developed at a rapid rate.

Conclusion

Three different approaches were used to address the same research

question, a survey, content analysis and web analytics. Each has its
own affordances and constraints, which have been briefly outlined
above. In particular, content analysis of weblogs and forum
transcripts revealed the best indication of student commitment to
CSCL. However, content analysis is notoriously difficult to perform
and the reliability of findings is often open to debate with general
agreement that the technique is “difficult, frustrating and time
consuming” (Garrison & Anderson, 2005, p. 132). All the same, the
findings of content analysis provided a rich array of information that
can help to inform future development of online teacher teams
(communities of practice). Perhaps the key observations identified
in the qualitative strand of the study were that online communities
of practice amongst UBD students need to be well structured,
moderated, focused and that COPs take time and moderator skill to
develop. This was to be expected and is supported by COP literature
(Lave & Wenger, 1991; Wenger, 1998; Wenger et al., 2002). Taking
this into account it may be that there is insufficient time in a one-
semester course to develop the skills and knowledge students must
have if they are to be expected to commit to CSCL. This study
seems to support this view. However, this problem can be
somewhat mitigated by ensuring that there is faculty wide support
for CSCL rather than limiting development of COPs to educational
technology courses. If CSCL is considered important and
worthwhile, because it is inevitably going to be a standard feature of
ongoing professional development of teachers and because there is
increasing focus on the need for teachers to work in teams, then
there is a better chance that students will get a consistent message
about the affordances of technologies within all the courses they
take.

The WEBLEI analysis provided a snapshot of participant

perceptions of their CSCL environment. The WEBLEI findings were
largely in agreement with content analysis findings. However,
quantitative analysis of web server logs revealed that there was a
difference between what participants believed and the actions they
took. On the one hand participants indicated that they were
satisfied with the results of online participation in an online
community and yet, for many participants, there is little evidence in
server logs to indicate that they were participating fully in their
online communities of practice. These contradictory findings may
be explained by the nature of the environment in which participants
had to work. They clearly valued participation and wanted to
collaborate online, but had insufficient time, skills or knowledge to

17
develop the social/communication aspects of an online community
of practice.

If CSCL is to work best then it requires participation from all

members of a team of teachers. Therefore it remains a concern
that levels of commitment were not uniform and that some
participants clearly did not commit to CSCL. Thus, it is important to
try to understand those factors that prevented some participants
from forming groups that collaborated successfully while others
foundered.

In summary, based on the amount of activity in sites, it is

possible to say that approximately 40% of participants were able to
generate some data to support the notion that they were fully
committed to CSCL. Of course others were also committed but were
unable to provide evidence of this because of the vagaries of the
environment in which they were operating. Still, there were others
that appear to have not been committed to CSCL, and the barriers
preventing full participation in CSCL need to be identified and
mitigated.

Recommendations
This research highlighted the need to learn more about the common
characteristics of successful CSCL teams, and the factors that prevented
some teams from being successful in a COP–CSCL environment. The study
showed CSCL environments to be complex with many interrelated factors
contributing to COP effectiveness, making them somewhat problematic to
implement. Therefore, having identified the need to develop a well
structured CSCL community development programme it remains to define
exactly what this will contain. The study results revealed that there
wasn’t sufficient time in a single semester course to develop the skills,
knowledge and propensities that all participants will need in their future
career and that a possible remedy to this problem would be to generate
faculty wide support for CSCL at all levels of preservice training.

References
American Association of Colleges for Teacher Education, & Committee on
Innovation and Technology. (2008). Handbook of technological
pedagogical content knowledge for educators. New York: Routledge.
Angeli, C. (2008). Distributed cognition: A framework for understanding
the role of computers in classroom teaching and learning. Journal of
Research on Technology in Education, 40(3), 271–279.
Bryman, A. (2008). Social research methods. Oxford: Oxford University
Press.
Carugati, F., Selleri, P., Matteucci, M. C., Tomasetto, C., Mazzoni, E., &
Gaffuri, P. (2008, September 1-2). Monitoring and analyzing
collaboration in e-learning environment: Two case studies applied
to vocational training. Paper presented at the Conference on
Knowledge Construction in E-learning Context: CSCL, ODL, ICT and
SNA in Education, Cesena, Italy

18
Chan, J. C. C., Hew, K. F., & Cheung, W. S. (2009). Asynchronous online
discussion thread development: Examining growth patterns and
peer-facilitation techniques. Journal of Computer Assisted Learning,
25(5), 438-452.
Chang, V., & Fisher, D. L. (2003a). The validation and application of a new
learning environment instrument for online learning in higher
education. In M. S. Khine & D. L. Fisher (Eds.), Technology-rich
learning environments: A future perspective (pp. 1-20). Singapore:
World Scientific Publishing Co. Pte. Ltd.
Chang, V., & Fisher, D. L. (Eds.). (2003b). The validation and application of
a new learning environment instrument for online learning in higher
education Singapore: World Scientific.
Creswell, J. W., & Plano Clark, V. L. (2007). Designing and conducting
mixed methods research. Thousand Oaks, California: Sage
Publications, Inc.
Dougimas, M., & Taylor, P. (2002). Interpretive analysis of an internet-
based course constructed using a new courseware tool called
Moodle. Paper presented at the 25th HERDSA Annual Conference.
Garrison, D. R., & Anderson, T. (2005). E-Learning in the 21st Century: A
Framework for Research and Practice. New York, NY:
RoutledgeFalmer.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory:
Strategies for qualitative research. Chigago: Aldine Publishing
Company.
Hughes, J. (2005). The role of teacher knowledge and learning experiences
in forming technology-integrated pedagogy. Journal of Technology
and Teacher Education, 13(2), 277-303.
Kopp, B., & Mandl, H. (2008, September 1-2). Collaborative learning: A
heterogeneous phenomenon. Paper presented at the Conference on
Knowledge Construction in E-learning Context: CSCL, ODL, ICT and
SNA in Education, Cesena, Italy
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral
participation. New York: Cambridge University Press.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA:
Sage.
Loucks-Horsley, S., Love, N., Stiles, K. E., Mundry, S., & Hewson, P. W.
(2003). Designing professional development for teachers of science
and mathematics. London: Corwin Press Inc.
Loucks-Horsley, S., & Matsumoto, C. (1999). Research in professional
development for teachers of mathematics and science: Research
into practice. School Science and Mathematics, 99(5), 258-271.
Mishra, P., & Koehler, M. J. (2006). Technological pedagogical content
knowledge: A framework for teacher knowledge. Teaches College
Record, 108(6), 1017-1054.
Moroney, M., Koay, T., & Boorer, D. (2009). Actual and preferred
perceptions of students engaged in collaborative online learning.
Journal of Applied Research in Education (In Press).
Salmon, G. (2005). E-Moderating: The key to teaching & learning (2nd
ed.). London: RoutledgeFalmer.
Soller, A., & Lesgold, A. (2007). Modeling the process of collaborative
learning. In H. Ulrich Hoppe, H. Ogata & A. Soller (Eds.), Computer-
supported collaborative learning: Studies in technology enhanced
collaborative learning New York: Springer Science + Business
Media.

19
Songer, N. (2007). Digital resources versus cognitive tools: A discussion of
learning science with technology. In S. K. Abell & N. G. Lederman
(Eds.), Handbook of research on science education (pp. 471-491).
New York: Routledge.
Stahl, G. (2004). Building collaborative knowing: Elements of a social
theory of CSCL. In J. Strijbos, P. Kirschner & R. Martens (Eds.), What
we know about CSCL and implementation in higher education (pp.
53-85). Boston: Kluwer Academic Publishers.
Tashakkori, A., & Teddlie, C. (2009). Foundations of mixed methods
research: Integrating quantitative and qualitative approaches in the
social and behavioral sciences. Thousand Oaks: Sage Publications
Inc.
Timperley, H., Wilson, A., Barrer, H., & Fung, I. (2007). Teacher
professional learning and development: Best evidence synthesis
iteration (BES). Auckland: Auckland University.
Wenger, E. (1998). Communities of practice: Learning, meaning, and
identity. . New York: Cambridge University Press.
Wenger, E., McDermott, R., & Snyder, W. (2002). Cultivating communities
of practice: A guide to managing knowledge. . Boston: Harvard
Business School.

About the author

Mike Moroney is an ICT/Educational Technology lecturer at the University

of Brunei. His research interest is in repurposing Internet moderated
communities of practice for inservice teacher professional development.
He is a PhD candidate at Auckland University.

Presented at ….

The 13th UNESCO-APEID

International Conference on Education
and World Bank-KERIS High Level Seminar
on ICT in Education
15-17 November 2009
Hangzhou, People’s Republic of China

20