The Granularity of Collaborative Work for Creating Adaptive

Learning Resources

Dade Nurjanah

Telkom School of Engineering, Telkom University, Jl. Telekomunikasi 1, Bandung, Indonesia

Keywords: Collaborative Authoring, Learning Designs, Pedagogical Knowledge, Asynchronous, Authoring

Granularity, Notes, History.

Abstract: Recent developments in the field of learning systems have led to adaptive learning which considers learner

models when performing pedagogical decisions. Problems emerge in providing knowledge spaces of

adaptive learning systems. As a knowledge space consists of pedagogical model, learner model, and

adaptation model, teachers need much effort to create it. This paper focuses on the authoring of the

knowledge spaces of adaptive learning systems and proposes a collaborative authoring approach for creating

pedagogical, learner, and adaptation models. The proposed approach combines asynchronous collaborative

work with Notes and History to support implicit coordination and workspace awareness. It applies IMS

Learning Design to represent the aforementioned models. To validate it, qualitative and quantitative

experiments were conducted. The experiment results indicated the high granularity of authoring, which

means that learning designers can efficiently and effectively work in an asynchronous collaborative

environment with Notes and History.

1 PROBLEMS IN AUTHORING

FOR ADAPTIVE LEARNING

Learning is a process to build knowledge and

enhance skills through studies, practices,

experiences, social interaction, lectures, or tutorials.

With many students registering in a course, teachers

are faced with various learners’ characteristics

differs. To accommodate the diversity, recent

developments in the field of learning systems have

led to adaptive learning which considers learner

models when performing pedagogical-related

decisions.

Along with its advantages, adaptive learning

system gives teachers or learning designers a

consequence to prepare a sheer sized and complex

learning space, consisting of domain, pedagogical,

learner, and adaptation models. Hence, it is difficult

for just one or two teachers to develop such a space.

Teachers need to work collaboratively to reduce

individual effort. Although teachers can work

individually on preparing courses, they should team

up with other teachers to check material consistency

and reliability, or to maintain learning resources

which are not fixed at certain stages, and to be kept

continuously updated.

A very common collaboration among teachers or

learning designers is on creating and reusing

learning content. It rarely happens on creating

pedagogical knowledge regarding how learning

content is delivered. This is contrary to the premise

suggesting that learning must be socially developed

(McDaniel and Colarulli, 1997). The collaboration

of learning designers involves multiple dimensions

(pedagogical, social, disiplinary, competency,

cultural, et cetera) which potentially improve

learning and benefit learners. Learning designers

themselves can get advantages from the

collaboration as they can learn new knowledge on

respective fields from their colleagues.

The collaboration, however, potentially fails

when learning designers can not gain concensuses

on various pedagogical preferences (Eisen and

Tisdell, 2013). Considering the potential advantages

and the possible failure of learning designer

collaboration, this paper discusses our study on the

collaborative work for authoring adaptive learning

resources. The study is motivated by a basic

question whether learning designers can or cannot

collaboratively work on authoring pedagogical,

learner, and adaptation models.

In this paper, we propose a collaborative work

165

Nurjanah D..

The Granularity of Collaborative Work for Creating Adaptive Learning Resources .

DOI: 10.5220/0004722701650173

In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 165-173

ISBN: 978-989-758-020-8

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

model for authoring learning designs. In the rest of

this paper, we discuss former studies on computer-

supported collaborative work (CSCW) and IMS

Learning Design (IMS LD) which is the only

learning standard supporting adaptation and

personalisation. Afterwards, research questions,

experiments, and data analysis results are described.

The contribution of this paper is presented in the

form of a demonstration showing that learning

designers can efficiently and effectively work in an

asynchronous collaborative environment with Notes

and History for creating adaptive learning resouces

represented in IMS LD.

2 THEORETICAL BACKGROUND

This paper concerns two issues in authoring learning

designers: learning standards to represent adaptive

learning resources and computer-supported

collaborative work to be applied.

2.1 Computer-Supported Collaborative

Work for Learning

CSCW has been successfully applied in various

areas for authoring various objects, such as

hypermedia documents (Haake, 1993), courseware

(Dicheva et al., 2002; Ras et al., 2008), academic

writing (Dimitrova et al., 2008), papers (Liccardi et

al., 2007), and ontology (Noy and Tudorache, 2008).

CSCW in particular enables social collaboration and

evolves knowledge on a large scale. It reduces

individual efforts, provides different insights, and

enhances the quality of output by enabling authors

from different expertise to work together (Noël and

Robert, 2004). Multiple persons who collectively

contribute their thoughts could surpass the

achievements of someone who works individually

(Dicheva et al., 2002; Posner and Baecker, 1992).

However, collaborative work may potentially

generate less positive output than individual work.

This would be more likely to be the case when

inappropriate communication and coordination

mechanisms are applied or workspace awareness is

limitedly supported (Gutwin and Greenberg, 2002;

Kittur et al., 2009; Lowry et al., 2005).

Communication and coordination methods

applied in online authoring are different from those

applied in traditional collaboration. In a traditional

collaboration, careful planning is important. It is

supported by face-to-face meeting, which is

beneficial to the authors as it offers interactive and

direct communication. In contrast, a careful plan is

not considered necessary in an online collaboration

where contributors have the freedom to do what they

consider important. Until recently, there have been

numerous research studies into how communication

mechanisms affect the authoring process and output.

It was found that the proper use of communication

method could improve the quality of artefacts (Kittur

and Kraut, 2008).

Workspace awareness is important for managing

coupling between working alone and working

together, simplifying communication, coordinating

actions, anticipating other authors’ actions, and

assisting authors (Gutwin and Greenberg, 2002).

Workspace awareness must be maintained, not only

in synchronous collaborative work, but also in

asynchronous collaborative work. Research on

workspace awareness in asynchronous collaborative

authoring was carried out with the same motivation

as in synchronous collaborative authoring (Dourish,

1997). Workspace awareness in asynchronous

collaborative work is related to the history of

occurrences, including actions, artefacts, events, and

authors’ presence and locations (Gutwin and

Greenberg, 2002).

2.2 IMS LD Supports for Adaptive

Learning

Learning design is motivated by a pedagogic

consideration that learning is not merely about a set

of learning objects, or simply content to be presented

to learners, but learning is also more about how the

materials are delivered to learners and how learners

can gain knowledge. People learn better if they are

actively involved in learning processes (Bonwell and

Eison, 1991). Hence, learning is carried out

according to a flow of learning activities, called

learning design, which consists of a structured set of

learning activities to be done by learners and support

activities to be carried out by teachers.

The need for learning design standards emerges

along with requirements to keep learning designs

consistent for all students. In addition, the use of

technologies for learning has raised the need for

reusable and interoperable digital learning designs.

Learning design standards, such as IMS SS and IMS

LD (Grocott et al., 2012), present some advantages

as they have well structures and abilities to include

learning objects as materials in order to support

lessons or learning activities.

In term of adaptive learning, IMS LD offers

wider adaptation and personalisation than IMS SS. It

supports flow-based adaptation, content-based

adaptation, and interactive problem solving-based

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

166

adaptation (Kravcik et al., 2008). ‘Hide’ and ‘Show’

are applied to lessons and activities for flow-based

adaptation and to resources for content-based

adaptation. Furthermore, they are applied for

adaptive problem solving assistance which is an

extension of flow-based adaptation. It provides

incremental-adaptive assistances, for example, by

applying time and/or the number of remediation.

The structure of IMS LD is presented in Figure 1.

title

learning-objectives

pre-requisites

components

roles

activities

learning-activity*

support-activity*

activity-structures*

environments

method

play*

act*

role-parts*

metadata

Figure 1: The structure of IMS LD (Grocott et al., 2012).

There has been an authoring tool of IMS LD. It is

called ReCourse (http:// tencompetence-

project.bolton.ac.uk/ldauthor/), that provides

functionalities for authoring and visualizing IMS

LD. In addition, there is CopperCore for validating

and delivering IMS LD.

3 RESEARCH QUESTIONS

Regarding the organisation of IMS LD which is

hierarchically structured, asynchronous collaborative

authoring with implicit coordination was considered

suitable. Former research has proved that implicit

coordination is more suitable for hierarchical tasks

or documents rather than explicit coordination

(Lowry, 2002; Lowry et al., 2005). Hence, we

hypothesise that learning designers do not need

intensive communication for coordination. The

hierarchical structure of IMS LD will make

authoring task division and assignment not too

complicated. In addition, standard meanings and

formats for all types of adaptive learning artefacts in

IMS LD will prevent a learning designer from

missunderstanding other authors’ work.

To test the suitability of the proposed method,

two experiments were conducted. They addressed

two research questions.

Question 1. With IMS LD that is hierarchically

structured, in which level of granularity is the

collaborative authoring carried out?

The proposed collaborative method is suitable for

authoring IMS LD if learning designers can

collaboratively work on the high and low levels of

learning designs, and also on adapting materials.

Accordingly, the observation identified the

contribution of authors in authoring three kinds of

pedagogical elements:

1. Plays and acts. Plays and acts are IMS LD

elements for pedagogical knowledge. Learning

designers’ contribution in authoring these

elements indicates that they can work

collaboratively on the high level of pedagogical

knowledge, which means that the granularity of

authoring is low.

2. Activities and Role-parts. In designing role-parts,

learning designers have to assign learning roles to

activities. Updates on learning activities, support

activities, activity groups, and role-parts indicate

that learning designers can work collaboratively

on the low level of pedagogical knowledge. It

means that the granularity of authoring is high.

3. Properties and conditions. Participants’

contribution in authoring properties and

conditions indicates that they can work

collaboratively on adapting materials. An

example of conditions is presented in Figure 2.

<imsld:conditions>

<imsld:if>

<imsld:not> <imsld:or>

<imsld:no-value>

<imsld:property-ref ref="P-completion-test-

advising"/>

</imsld:no-value>

<imsld:no-value>

<imsld:property-ref ref="P-completion-test-

anticipating"/>

</imsld:no-value>

</imsld:or> </imsld:not>

</imsld:if>

<imsld:then> <imsld:show>

<imsld:learning-activity-ref ref="LA-request-

grade"/>

<imsld:environment-ref ref="E-background"/>

</imsld:show> </imsld:then>

</imsld:conditions>

Figure 2: An example of adaptation rule in IMS LD.

Question 2. Are Notes suitable for implicit

coordination and, with History, suitable for

workspace awareness in collaborative authoring of

IMS LD?

Experiments to answer this question would refer to

former studies on CSCW which have confirmed that

coordination mechanisms are group sized-specific.

TheGranularityofCollaborativeWorkforCreatingAdaptiveLearningResources

167

They have examined the influence of the number of

contributors and the independency of collaboration

tasks (Kittur et al., 2009). Kittur and Kraut (Kittur

and Kraut, 2008) highlighted the correlations

between implicit coordination, early stages of

authoring, and the quality of articles. The advantages

of implicit coordination are greater during the early

phases of authoring, when the article is in its earliest

versions. During these phases, outlining the article

structure by a subset of authors will lead to greater

increases in quality. When the authoring work is

carried out by the small subset of authors, the quality

of articles will increase and is better than articles

produced by group where the work is evenly divided

amongst all authors.

4 THE EXPERIMENTS

Two experiments were conducted to answer the

research questions: a qualitative inquiry with

observation and interview and a between-group

quantitative inquiry with questionnaires. The

proposed collaborative authoring model was

implemented by extending ReCourse, a stand-alone

open sourced tool, The main functionalities of

ReCourse can be classified into five groups:

1. Manage domain model implemented in

resources.

2. Manage goal model implemented in learning

objectives, pre-requisites, course overview, role,

plays and acts, learning and support activities,

activity groups, role-parts, conditions, and

environments.

3. Manage learner model implemented in global-

personal properties for learners’ profiles and

local-personal properties for learners’ progress.

4. Manage adaptation model implemented in pre-

defined and user-defined conditional rules.

5. Validate learning designs.

For the experiments, ReCourse was extended

with supporting functions for collaboration. The new

functionalities in Collaborative ReCourse consist of

(Nurjanah, 2013):

1. User group management. The first author is

assigned as the coordinator who has an authority

to add new members into the group; the others

are called members. These are the only role

assignments in the proposed authoring method.

2. Notes. Notes were provided in three types based

on the types of comments possibly posted by

learning designers. First, Note is attached to the

whole learning design. It is provided for learning

designers to share comments about the learning

design itself, learning objectives, pre-requisite

courses, completing rules, or other general

comments. Second, Note is attached to History,

called History’s Note, which is aimed to

maintain learning designers’ comments

regarding updates they made. Third, Notes are

attached to IMS LD elements, called objects’

Notes. One object’s Note attaches to one play,

act, learning or support activity, activity group,

property, condition, role, role-part, or resource.

Objects’ Notes aim to maintain authors’

comments regarding particular elements. The

Observation investigated which type(s) of Notes

participants prefer.

Figure 3: A screenshot of Collaborative ReCourse prototype

with objects’ Notes.

3. History, a feature to record provenance

information about changes, the types of changes,

the affected objects, and the learning designers

who made the changes.

4. Existing learning content gallery. This is an

additional feature in which authors can select,

add, or tag learning materials. This feature aims

to decrease authors’ effort when creating

learning content and to enhance authors’

awareness of the availability of learning

materials to be reused.

The architecture of Collaborative ReCourse

prototype can be found in Figure 4.

4.1 Qualitative Inquiry: Observation

and Structured Interview

This experiment aimed to observe the granularity of

collaborative authoring of learning designs. It

investigated how learning designers did

collaborative work and on which elements the

collaboration was carried out. It investigated

whether they could collaboratively work only on the

top level of pedagogical resources (plays and acts)

or on the low level (role-parts) as well. Furthermore,

it observed authors’ contribution in authoring

adapting materials (conditions and properties). At

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

168

Figure 4: The architecture of Collaborative ReCourse.

the end, a structured interview was conducted after

the observation to gather participants’ opinion about

the authoring process and the collaboration features:

Notes and History.

As this is a qualitative experiment, a limited

number of participants were required; too many

participants will lead to divergent results (Marshall

and Rossman, 2006). There were 12 people

participating in the experiment. They were recruited

by personal email invitation. To select participants,

purposeful sampling as opposed to random sampling

was used. Participants were selected by considering:

 Gender. Since collaborative authoring of

learning designs is not gender specific, male and

female participants in a balanced composition

were involved in this experiment.

 Teaching experience. Participants were those

who had teaching experience in classrooms or

laboratories.

 Java knowledge. It was needed because

participants were required to develop a learning

design of Java programming.

Observation. All participants were assigned to work

in asynchronous collaborative authoring

environments. They were divided evenly into four

groups. In the observation, participants 1 to 3

worked as group A, participants 4 to 6 worked as

group B, participants 7 to 9 worked as group C, and

participants 10 to 12 worked as group D. Each group

was required to create a learning design of Java

programming in nine sessions of 60 minutes. Each

participant was required to work in three non-

consecutive sessions. There was no authoring

scenario to be followed by participants; they were

free to make any update.

All participants worked in the similar environment:

asynchronous collaboration. The only difference is

that group C and D were supported by workspace

awareness features in the forms of Notes. They

could communicate through Notes and access

provenance information (History), such as what

recent updates that have been made, by whom and

when. Such features were disabled for group A and

B. Although participants worked collaboratively, the

focus of the observation is individual actions in the

collaborative work.

Results. The results obtained from the observation

were presented in the following graphs. First, the

contribution of authors in authoring the

aforementioned three kinds of pedagogical elements

is presented in Figure 5. As shown in the graph, all

participants contributed in the authoring. However,

there were two participants in group A and B,

participants 2 and 4, dominant over the others in

their own groups.

Figure 5: Participants’ contribution in authoring all

pedagogical resources.

Second, we broke down the data to see the

granularity level of authoring which is indicated by

the contribution of authors in authoring learning

activities and role-parts. As we have discussed, the

contribution of authors in authoring learning

activities and role-parts indicates the high

granularity level of authoring. As shown in Figure 6,

all participants participated in authoring learning

activities and role-parts in various contribution. Like

in the previous finding, there were participants who

contributed more than fifty percents in group A and

The last focus of the observation is authoring

learner model in the form of properties and adapting

elements which consist of predefined- and user-

defined conditions. As shown in Table 1, all

participants contributed in the authoring. However,

participants 7 to 12 supported with Notes and

History presented better contribution as there were

not properties and neither rules which were

individually authored by sole participants.

17,81%

57,08%

25,11%

72,15%

15,82%

12,03%

43,70%

23,53%

32,77%

41,98%

29,01%

20%

40%

60%

80%

123456789101112

TheGranularityofCollaborativeWorkforCreatingAdaptiveLearningResources

169

Figure 6: Participants’ contribution in authoring activities

and role-parts, the lowest level pedagogical resources.

Table 1: Participants’ contribution in authoring learner

model and adaptation rules.

Participants

Learner model and adapting materials

Properties

Predefined

Rules

User-defined

Rules

1 25.0% 100.0%

2 60.0%

3 75.0% 40.0%

4 81.0% 100.0%

5 100.0%

6 19.0%

7 16.7% 50.0% 14.3%

8 50.0% 50.0% 57.2%

9 33.3% 28.5%

10 33.3% 50.0% 18.0%

11 40.0% 64.0%

12 66.7% 10.0% 18.0%

Discussion. The observation has investigated how

learning designers work in asynchronous

collaborative enviroments with features for limited

communication (Notes) and awareness supports

(Notes and History). It revealed a fact that learning

designers can work collaboratively in authoring all

IMS LD elements. The granularity of authoring is

high as they can work collaboratively from plays to

role-parts and from non-adapting to adapting

materials. In term of the usability of Notes, the

observation shows that among the aforementioned

three kinds of Notes, History’s Note is the least

accessed one. Participants prefer to use Note and

objects’ Notes as they thought that the function of

History’s Note has been covered in Note.

4.2 Quantitative Inquiry: Between-

Group Questionnaires

IMS LD offers advantages for adaptation and

interoperability. IMS LD, however, does not provide

an element or any space for learning designers to put

notes or comments, such as to explain what the

objectives of learning activities, why a particular

topic is important, et cetera. We proposed Notes to

enable learning designers to put comments regarding

the authoring process or the authored artefacts and

History that, with Notes, describes how the

authoring process is going on. The second

experiment aimed to investigate whether Notes and

History give positive impacts in authoring IMS LD.

Method. Adaptation model is one component of

adaptive learning resources that is considered to be

more difficult to understand than other resources. In

this study, a comparison between Group 1 and

Group 2 was drawn to see if implicit coordination

and workspace awareness features is suitable for

authoring adaptive learning resources. Both groups

are assigned to work in asynchronous collaborative

environments, but Group 2 was supported with

features for communication and workspace

awareness, while Group 1 was not. It could be

concluded that Notes and History give positive

impacts to authoring, if Group 2 presented better

workspace awareness than Group 1.

There were 44 participants who participated in

the experiment. The number of participants was

estimated by G*Power software (Hendrix et al.,

2008). They had teaching experience and Java

knowledge. Like in the first experiment, they were

required to involve in collaborative authoring of

learning designs in asynchronous-collaborative

environments.

Participants were divided into two groups. One

group was supported with Notes and History, while

the other group was not. To guarantee that

participants have the same profiles regarding their

teaching experience, IMS LD authoring experience,

and Java knowledge, we conducted a MANOVA test

to see if there is a significant difference between the

groups. The test comprising Pillai's Trace, Wilks'

Lambda, Hotelling's Trace, and Roy's Largest Root

confirmed that there is no significant difference

between the profiles of Group 1 and Group 2.

Table 2. The MANOVA results for participants’ profiles.

Effect Value F df Error df Sig.

Pillai's Trace

Wilks' Lambda

Hotelling's Trace

Roy's Largest Root

.244

.756

.322

2.449

5.000

38.000

.051

Contrary to the first experiment, this experiment

required participants to follow artificial authoring

scenarios. To give participants knowledge about

IMS LD and ReCourse, we arranged a 45-minute

introduction session that all participants had to

attend. In this session, participants were free to

explore the tool and examples of IMS LD.

The questionnaires were reviewed by two senior

18,97%

55,17%

25,86%

53,85%

20,51%

25,64%

37,50%

28,13%

34,38%

45,65%

32,61%

21,74%

10%

20%

30%

40%

50%

60%

123456789101112

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

170

experts and one junior expert from related fields.

The aim of this review was to ensure that targeted

information could be gained through all the

questions, and to avoid ambiguity of words in the

questionnaires that possibly cause misunderstanding.

All reviewers had similarity profiles with

participants that they have teaching experience as

well as educational backgrounds in engineering. As

this experiment was carried out not only in the UK,

but also in Indonesia, it was essential that at least

one reviewer was fluent in both Indonesian language

and English.

Results. All participants were required to find

information in a predefined unit of learning of Java

Programming. They were free to explore the UoL.

There was no guidance given to Group 2 participants

as to where to find notes written by previous

learning designers. Afterwards, all participants were

asked a number of questions related to adapting

materials. The questions covered five cases; each

case employed simple rules and logic that learning

designers could easily follow. Participants were

required to observe the case and answer questions

related to the case. One example of the questions is

presented below:

Please find rules. You will find one rule:

“Rule 1”. What is the objective of the rule?

Participants’ answers indicated their workspace

awareness. The questions used three nominal values

to classify users’ answers: wrong answers, no

answers, and correct answers. A comparison

between the number of correct answers given by

Group 1 and Group 2 is described in Figure 7

(Nurjanah and Davis, 2012). In each case, Group 2,

which supported with Notes and provenance

information, gave a higher percentage of correct

answers than Group 1.

Figure 7: A comparison of users’ understanding between

Group 1 and Group 2.

Further study was carried out to Group 2. The

same approach was conducted to Group 2 in

authoring two other courses: Introduction to Biology

and Web Programming. A classification of authoring

processes were applied to find out whether the

proposed authoring approach is suitable only for a

particular stage or for all stages. Authoring Biology

and Web Programming were in early stages of

authoring, while authoring Java Programming was in

an advance stage. The experiment result shows that

the proposed authoring approach is suitable for both

early and further stages of authoring.

Figure 8: Participants’ awareness.

Discussion. In the second experiment, learning

designers were required to make some updates in

ongoing collaborative authoring. They were required

to understand how the authoring was going on.

Group 1 could gain awareness only from the curremt

states of the authored learning designs, while Group

2 could also learn from the provided Notes and

History. The experiment result presents an evidence

that Notes and History give positive implication to

learning designers’ awareness in early and further

stages of authoring.

5 CONCLUSIONS

The granularity of authoring indicates that implicit

coordination is appropriate for collaborative

authoring of IMS LD. The data analysis results

showed that participants worked collaboratively in

authoring pedagogical knowledge, including

adapting materials. The granularity of authoring is

high since they did collaboration in authoring all

IMS LD elements, including plays and the

underlying elements (acts and role-parts),

learning/support activities and activity groups,

properties, and conditions. As former studies on

adaptive learning have proved that people can work

collaboratively in authoring learning content, this

experiment confirms that they also can

collaboratively work in creating pedagogical

knowledge (Conclusion 1).

Second, the usability of Notes and History was

tested through a between-group quantitative study.

The study compared the workspace awareness and

25,45%

42,73%

31,82%

9,09%

10,91%

80,00%

Gavewronganswers Gavenoanswers Gavecorrectanswers

Group1 Group2

92,42%

3,03%

4,55%

80,81%

8,59%

10,61%

20%

40%

60%

80%

100%

Gavecorrect

answers

Gavenoanswers Gavewrong

answer

Earlystagesofauthoring Furtherstagesofauthoring

TheGranularityofCollaborativeWorkforCreatingAdaptiveLearningResources

171

the contribution of two groups of learning designers;

one group was supported with Notes and History,

while the other was not. Learning designers

supported with Notes and History presented better

contribution and higher workspace awareness than

the others. They understood what updates had been

made, what other authors had done, what the reasons

of updates, and who made update. To conclude, the

experiment results present evidence that Notes and

History give positive impacts in authoring IMS LD

(Conclusion 2). Another finding was about the

importance of Notes. Learning designers considers

that Notes and objects’ Notes are more necessary

than History’s Note (Conclusion 3).

Finally, although the experiments have

confirmed that asynchronous collaborative authoring

method with features for limited communication

(Notes) and workspace awareness (Notes and

History) is suitable for authoring learning designs,

further studies to compare this approach with other

approaches are required to find the best approach for

collaborative authoring of IMS LD.

REFERENCES

Bonwell, C. & Eison, J. A., 1991. Active Learning:

Creating Excitement in the Classroom, Jossey-Bass.

USA, 1st edition.

Dicheva, D., Aroyo, L. & Cristea, A., 2002. Collaborative

Courseware Authoring Support. In the proceeding of

the International Conference on Computers and

Advanced Technology in Education, pp. 52-57. ACTA

Pres.

Dimitrova, V., Lau, L. & Le Bek, A., 2008. Sharing of

Community Practice through Semantics: A Case Study

in Academic Writing. In the proceeding of Sixth

International Workshop on Ontologies and Semantic

Web for E-Learning, pp. 30-39.

Dourish, P., 1997. Extending Awareness Beyond

Synchronous Collaboration. The Computer and

Human Interaction (CHI) conference. Los Angeles.

Eisen, M. J. & Tisdell, E. J., 2013. Team Teaching: The

Learning Side of the Teaching-Learning Equation.

Toward the Best in the Academy, vol. 14.

Grocott, S., Halm, M., Lefrere, P., Manderveld, J., Mason,

J., Norton, M., Ostyn, C., Rehak, D., Sloep, P.,

Springer, G., Taliesin, B., Todd, R., Towle, B.,

Vogten, H., Washburn, C. & White, B., 2012.

Learning Design Specification. From http://

www.imsglobal.org/learningdesign/. Accessed on

March 1, 2012.

Gutwin, C. & Greenberg, S., 2002. A Descriptive

Framework of Workspace Awareness for Real-Time

Groupware. Computer-Supported Cooperative Work,

vol. 11.

Haake, J. M., Neuwirth, C.M., 1993. Collaborative

Authoring of Hypermedia Documents. Machine

Translation Today, Translating and the Computers,

vol. 15.

Hendrix, M., Bra, P. D., Pechenizkiy, M., Smits, D. &

Cristea, A., 2008. Defining Adaptation in a Generic

Multi Layer Model. CAM: the GRAPPLE Conceptual

Adaptation Model. In the proceeding of EC-TEL 2008,

pp. 132-143. Springer.

Kittur, A. & Kraut, R. E., 2008. Harnessing the Wisdom

of Crowds in Wikipedia: Quality Through

Coordination. The 2008 ACM Conference on

Computer Supported Cooperative Work. San Diego,

CA, USA.

Kittur, A., Lee, B. & Kraut, R. E., 2009. Coordination in

Collective Intelligence: The Role of Team Structure

and Task Interdependence. The 27th International

Conference on Human Factors in Computing Systems.

Boston, MA, USA.

Kravcik, M., Burgos, D., Lebrun, M. & Oneto, L., 2008.

Integrated Model of Adaptive Learning Based on

Standards. In GRAPPLE Deliverable 5.1 version 1.0.

Liccardi, I., Davis, H. C. & White, S., 2007. CAWS: a

wiki system to improve workspace awareness to

advance effectiveness of co-authoring activities. CHI

'07 extended abstracts on Human factors in computing

systems. San Jose, CA, USA.

Lowry, P. B., 2002. Research on Process Structure for

Distributed, Asynchronous Collaborative Writing

Groups. The 8th Americas Conference on Information

Systems

. Dallas, USA.

Lowry, P. B., Nunamaker, J. F., Jr., Curtis, A. & Lowry,

M. R., 2005. The Impact of Process Structure on

Novice, Virtual Collaborative Writing Teams. Journal

of IEEE Transactions on Professional

Communication, vol. 48.

Marshall, C. & Rossman, G. B., 2006. Designing

Qualitative Research, Sage Publication. Thousand

Oak - London - Delhi, 4th edition.

Mcdaniel, E. A. & Colarulli, G. C., 1997. Collaborative

teaching in the face of productivity concerns: The

dispersed team model. Innovative Higher Education,

vol. 22.

Noël, S. & Robert, J.-M., 2004. Empirical Study on

Collaborative Writing: What Do Co-authors Do, Use,

and Like? Computer Supported Cooperative Work:

The Journal of Collaborative Computing, vol. 13(1).

Noy, N. F. & Tudorache, T., 2008. Collaborative

Ontology Development on The (Semantic) Web. The

AAAI Spring Symposium on "Semantic Web and

Knowledge Engineering" (SWKE). Stanford.

Nurjanah, D. & Davis, H. C., 2012. Improving the

Workspace Awareness of Authors in Asynchronous

Collaborative Authoring of Learning Designs. ED

MEDIA ’12. Denver - Colorado, USA.

Nurjanah, D., 2013. Awareness Support for Learning

Designers in Collaborative Authoring for Adaptive

Learning. Ph.D thesis, 202 pages. Southampton: The

University of Southampton.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

172

Posner, I. R. & Baecker, R. M., 1992. How People Write

Together [Groupware]. In the proceeding of the 25th

Hawaii International Conference on System Sciences,

pp. 127-138.

Ras, E., Rech, J. & Weber, S., 2008. Collaborative

Authoring of Learning Elements for Adaptive

Learning Spaces. The 5th International Conference on

Adaptive Hypermedia and Adaptive Web-Based

Systems. Hannover, Germany.

TheGranularityofCollaborativeWorkforCreatingAdaptiveLearningResources

173