PROJECT-BASED COLLABORATIVE LEARNING

WITH ARGUMENTS

Flexible Integration of Digital Resources

Gert Faustmann

Department of Cooperative Studies, Berlin School of Economics and Law, Alt-Friedrichsfelde 60, 10315 Berlin, Germany

Keywords: Project-based learning, Collaborative learning, Cscl, Arguments, Digital resources.

Abstract: Success in project-based learning strongly depends on the experience of the project participants.

Experienced learners often develop results that are not understood by less experienced participants. In this

situation arguments can show the solution’s approach by connecting parts of the given task with individual

results. By using digital documents of arbitrary format it is possible to build a flexible net of arguments for

the whole project. This paper describes an approach to integrate arguments with project-based learning and

shows how to realize the use of digital resources in a web-based learning platform.

1 INTRODUCTION

Project-based learning is a widely accepted approach

to teach and train problem solving capabilities.

Especially in higher education complex tasks should

be a subject in order to simulate and train real world

problems. They are open tasks in the way that on the

one hand the initial question may change according

to a later research process and on the other there is

not just a single solution (Jung et al., 2001). This

kind of task is particular suitable for universities,

because students can be prepared for today’s

business world’s requirements. Additionally the

integration of projects in class is motivating for the

students and considerably improves their

performance (Doppelt, 2003).

But some problems occur in realizing this

learning approach: when using project-based

scenarios in higher education, varying levels of

student knowledge often are evident. An example is

design and implementation of software systems in

early computer science classes. Students have

different prior knowledge according to their interests

and education in school. This results in different

competencies and volume of output. As a

consequence after some time working in a project

students with low experience in software

development cannot further contribute ideas to the

project, because they do not understand most of the

parts of the solution already completed.

This paper introduces a concept to compensate

different levels of prior knowledge by making

project decisions more comprehendable. Less

experienced project members will then be able to

follow the solution process. This support will be

implemented by arguments that combine parts of a

solution with the requirements formulated in the

project assignment or parts of it. An argument in this

case need not be a logical assertion or a rule, but is

information that helps to understand the particular

result. We implemented the web-based learning

environment WeCoLAr (We

b-based Collaborative

earning with Arguments) that allows storing

arbitrary digital documents as arguments and that

gives the opportunity to navigate through a clearly

arranged solution structure.

The following chapters at first introduce the

constructivistic approach in collaborative online

learning and show why and how to incorporate

arguments into collaborative learning environments.

Then the concept of arguments as supporting

knowledge in problem solving is developed.

Especially the integration of digital documents of

arbitrary format will be shown. The paper concludes

by discussing the presented approach and showing

possible extensions.

359

Faustmann G..

PROJECT-BASED COLLABORATIVE LEARNING WITH ARGUMENTS - Flexible Integration of Digital Resources.

DOI: 10.5220/0003483403590364

In Proceedings of the 3rd International Conference on Computer Supported Education (CeLS-2011), pages 359-364

ISBN: 978-989-8425-50-8

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

2 RELATED WORK

While collaborative learning based on

constructivistic principles is an accepted concept in

learning theory there are different ways how

arguments are used to support learning. Some

approaches exist how to present these arguments to

the students.

2.1 Constructivistic Learning

Approach

Today’s theory of learning favours a constructivistic

approach, in which the student uses his/her

experience for problem construction and problem

solving. The teacher takes the role of a coach

supporting the learners in managing the complex

situation (Hmelo-Silver, 2004). In (Gilbert and

Driscoll, 2002) constructivistic principles of learning

are defined by

1. A group objective,

2. A cooperative group with interactions,

3. Individual decisions and means,

4. The use of integrated tools for saving

results and for communication.

These principles are to be realized in tool-based

online learning environments facilitating distributed

learning. Especially wikis as tools to allow the

distributed creation of documents play an important

role (Augar et al., 2006). They are used in settings

with different learning objectives (e.g. journalism

(Ma and Yuen, 2008). Extending wiki tools for

learning scenarios is investigated by some authors

(e.g. in (Larusson and Alterman, 2009)(Pusey and

Meiselwitz, 2009)). Wiki platforms can therefore be

judged as an elementary part of online learning

plattforms.

2.2 Arguments in Collaborative

Learning

While wikis support a distributed collaborative

creation of documents they do not offer explicit

support in understanding the knowledge building

process. There is plenty of work on how to use

arguments to support project based learning.

(Scheuer et al., 2010) give a very detailed insight to

approaches and systems in this area. One common

goal is to support the akquisition of argumentation

skills in computer supported distributed learning

platforms. The research questions in this area are

− the visualization of argument structures,

− the interaction between students via their

arguments,

− the connection of arguments with ontologies

and,

− the analysis of arguments with feedback to the

students.

For the goal of balancing very different prior

knowledge of the learners particularly the

presentation (and with it the visualization) of the

arguments and the knowledge may be of interest.

2.3 Visualizing Arguments

There is concrete research on how to visualize

argument structures. Visualization supports the

awareness of the other participants’ knowledge and

leads to a more efficient generation of new

knowledge. (Sbarski et al., 2008) design a

visualization tool to support a better understanding

of argumentation structures. (Suthers et al., 2008)

carry out a study investigating the support of

knowledge maps in threaded discussions. (González

et al., 2007) construct dialectical trees representing

arguments automatically extracted from discussions.

An interesting software tool to incorporate and

visualize arguments is SenseMaker (Bell and Linn,

2000). Arguments are web resources that are

presented with their URLs. The overall structure is a

container style where arguments can contain further

arguments and so on.

(Land and Greene, 2000) investigate the

knowledge building process in project-based settings

from the perspective of information resources. They

observe a topic drift while learners search for new

information. Structuring the information is no

subject in their study.

3 LEARNING SUPPORT WITH

ARGUMENTS

3.1 Arguments as Explanations

Arguments in the proposed concept are used as

explanatory statements for project results. They

connect parts of the project assignment with parts of

the solution. Beyond this connection arguments have

a meaning that gives background knowledge for the

way the part of the solution they are pointing to is

designed.

Arguments need not be a logical statement

giving a proof for the combined result. They support

other project participants in understanding the result

and its relation to the task.

CSEDU 2011 - 3rd International Conference on Computer Supported Education

360

Figure 1: Arguments combining Tasks and Results.

In this way one can think of many types of

documents being an argument. Knowledge from

textbooks or lecture slides can explain a result.

Active knowledge showing a method to solve a

problem that is relevant for the given task gives

good support for understanding a result. Examples

for active knowledge documents are sample

solutions and newsgroup articles. Finally, it should

be possible to use arguments that are developed by

the author of a result herself/himself. This can be an

explanation in textual or graphical form.

3.2 Integration of Arguments

At this point we combine documents to show a

development from tasks to results. In practice it is

difficult to relate information on document level

with one another. A task provided in an individual

document may have different sub tasks. The same is

with a resulting document that may have different

solution aspects explained by different argument

documents (see Figure 1).

Let us look at the following example: The project

task is to develop a small software system that

provides a command line interface (CLI).

This interface offers commands that allow the

management of directories and files (e.g. deleting

files and directories, showing the content of a file,

making a new directory). Not only the program text

is a result of the project but also some graphical

design documents visualizing the algorithms used to

implement the different commands. Furthermore

there is a requirement that the interface should easily

be extended by further commands. These

assignments are formulated in one document.

Resulting documents can now be the whole program

code and graphical documents (e.g. in jpg format).

Argument documents can be documents about how

to design algorithms in a graphical way (e.g. basics

of Nassi-Shneiderman-diagrams), parts of textbooks

about programming basics and pattern documents

showing how to make software programs easily

extendable.

It is obvious in this example that linking whole

documents that only are partially relevant is not

useful. It should be possible to link just a piece of a

textbook (e.g. how to realize an iterating program

structure) or a sub task of the whole project task

(e.g. just the “extension requirement” to link it with

a command pattern document).

Figure 2 shows the relation between tasks,

arguments, results and documents in the conceptual

approach. Tasks, arguments and results are in a

PROJECT-BASED COLLABORATIVE LEARNING WITH ARGUMENTS - Flexible Integration of Digital Resources

361

direct relation with arguments combining tasks and

their results. They all three are parts of documents

which themselves belong to exactly one

document. The approach is to make a difference

between a whole document and the parts of this

document. Later on, a technical solution will be

presented to implement these document parts.

Note that every resulting document must at least

have one argument to be related to. This enforces the

development of a net of arguments that can explain

the whole set of results. But it is also possible that a

result has more than one explaining argument. In our

example it may be difficult to find the parts of the

resulting program text corresponding to documents

explaining iteration (for repeated reading of

commands) and string analysis (for analysing the

user’s input). So it should be possible in this

example to specify arguments pointing to the

reading routine and also pointing to the routine

analyzing the string typed in by the user.

Figure 2: Integration of document artefacts.

3.3 Collaborative Learning Process

with Arguments

The provision of arguments in the learning process

can be twofolded: On the one hand the author of a

new result in the project should present one or more

arguments for her/his particular result. On the other

the teacher may want to give some hints in the

beginning of the project by providing arguments that

may lead to results later on. This may also be done

by the teacher while the project is already taking

place. She/he is then able to lead the work of the

project participants in the intended direction.

Taking the example of a software development

project the teacher can provide the assignment

document to develop the CLI together with

documents describing the basic functionality of a

command line interface. Moreover she/he can

provide lecture slides showing some alternatives for

making a software system extendable (e.g. some

software patterns like command pattern or decorator

pattern). Then the learners work on the project by

adding results to the project space. Suppose the

project members provide the program text

implementing the input and analyzing routines and

three commands. They also contribute digrams to

show how the command routines are implemented.

They must also provide argument documents for

each result. An argument document can be a

document provided by the teacher or a document

that the author of the result loads into the learning

platform. While accompanying the project the

teacher may give hints how to implement the

software system. Suppose she/he looks at the

program text so far and is not satisfied with the

solution. Then she/he may load a sample

implementation of a command pattern into the

project space to be used by the participants as an

argument for their own implementation of an

extendable system.

4 USING DIGITAL RESOURCES

This chapter proposes a concept that allows to

implement document parts of arbitrary format.

Especially the problem of a simple and ergonomic

use of these document parts is discussed.

4.1 Format Heterogeneity

When trying to use portions of documents in a

learning environment some problems appear: Many

types of documents can be arguments within the

suggested system approach. So there also may be

very different document formats that have to be

integrated in the system. The problem of

representing parts of a document is even worse

under the condition of managing multiple formats.

Besides it should be possible for every project

member to work at the documents and e.g. extend

them. This means that every document must be

available in its native format (e.g. the format of the

used text processing system). At least there should

be no redundance if a document is used in many

contexts in a project.

The approach to deal with these requirements is

visualized in Figure 3. The system stores not only

the original document but also a universal

representation of it. This will then allow to use a

general method to handle the documents in the user

interface (viewing, moving, etc.) and to work with

parts of a document.

Each native document within the proposed

CSEDU 2011 - 3rd International Conference on Computer Supported Education

362

Figure 3: Realizing Document Parts.

system has a representation in a graphical format

(here jpg). This is a 1:1-relation meaning that a

document presented on the graphical user interface

(the jpg-document) can easily be loaded in its native

format. The graphical document can then easily be

marked to define sub parts of the whole document.

One document may have many markings. A marking

just denotes a rectangular area in the graphical

document by storing two points defining this

rectangular area. Considering the argument net in

which tasks are combined with results using

arguments one can see from this implementation

view that the nodes in the net are just markings

related to graphical documents. On the other side,

comparable to a document pool, there are the

original documents and the graphical representations

of them.

4.2 Overview and Navigation

While other approaches combine the graphical

argumentation structure with text based arguments,

arguments in this paper often are based on a

graphical format (e.g. a lecture slide with some

graphical portion).

As a consequence the technical concept includes

the visualization of documents when presenting the

whole argument net to the user. This is supported by

the previously offered design rationale to present

documents by their graphical representations. So the

project member is able to see the dependencies

between results and sub tasks as well as the contents

of the documents respectively the parts of the

documents.

The requirement to work with the document

results in the approach to be able to have different

zooming levels to look at documents, to get a full

view of documents in the learning system and finally

to save the original document into a participant’s

local memory. Because a net of arguments can grow

rapidly over time it is possible to filter the

perspective acording to the structure.

Dimensions in filtering may be the author (e.g.

just see one’s own results), or results and arguments

based on one part of the assignment (e.g. show all

about realizing an extendable software system).

4.3 Learning Platform WeCoLAr

We implemented a first prototype in a web-based

platform for collaborative learning. The WeCoLAr-

system is programmed in PHP and AJAX for

building an interactive user interfaces. It is for

example possible to freely move documents within

the argument net to position new arguments.

Figure 4: WeCoLAr User Interface.

PROJECT-BASED COLLABORATIVE LEARNING WITH ARGUMENTS - Flexible Integration of Digital Resources

363

Figure 4 gives an idea of the user interface of

WeCoLAr.

5 SITUATION AND OUTLOOK

The presented approach is based on the

constructivistic idea of wikis in collaborative

learning environments (Augar et al., 2006). It

contrasts the so far use of arguments in assisting

discussions (e.g. in (Sbarski et al., 2008)) and

supports the understanding of project results of

arbitrary type and format. While SenseMaker also

uses arguments for collaborative problem solving, it

focuses on the logical argumentation process in

scientific problems (Bell and Linn, 2000).

WeCoLAr will be tested in real world settings at

university level. Beside the topic of programming

software systems, we will try to use the system in

business courses e.g. macroeconomics. Future

conceptual work will incorporate a scripting

approach to support the project members but also to

enforce everyone’s contribution to a project. These

scripts will enforce time aspects (e.g. working on

assignments one by one), and role assignment (e.g.

defining who has to do which part of a project). A

further control of the project work may also result in

new learning models (e.g. provide a solution and let

students find arguments for the solution). Further

research is the implementation of conditions for the

results of a project. Especially in software

engineering many documents are compiled that all

depend on one another. A problem for beginners

when learning software development is exactly this

dependancy between the different documents (e.g.

program code and design documents). We will make

these conditions also part of the learning process.

ACKNOWLEDGEMENTS

The project WeCoLAr is funded by the European

Social Fund (ESF 2009000038) as a sub project

within „E(r)lernen: Kompetenzvermittlung zum

Einsatz neuer Medien“. I like to thank Liangliang

Gu for his contribution to the WeCoLAr system.

REFERENCES

Augar, N., Raitman, R., Zhou, W., 2006. Wikis:

Collaborative Virtual Learning Environments. In The

International Handbook of Virtual Learning

Environments. pp. 1251-1269.

Bell, P., Linn, M., 2000. Scientific arguments as learning

artifacts: Designing for learning from the web with

KIE. International Journal of Science Education,

22(8), pp.797–817.

Doppelt, Y., 2003. Implementation and assessment of

project-based learning in a flexible environment.

International Journal of Technology and Design

Education, 13(3), pp.255–272.

Gilbert, N., Driscoll, M., 2002. Collaborative knowledge

building: A case study. Educational Technology

Research and Development, 50(1), pp.59-79.

González, M. et al., 2007. Modelling Shared Knowledge

and Shared Knowledge Awareness in CSCL Scenarios

Through Automated Argumentation Systems. In

Groupware: Design, Implementation, and Use. pp.

207-222.

Hmelo-Silver, C.E., 2004. Problem-based learning: what

and how do students learn? Educational Psychology

Review, 16(3), pp.235–266.

Jung, H., Jun, W., Gruenwald, L., 2001. A design and

Implementation of Web-Based Project-Based Learning

Support Systems. The Human Society and the Internet

Internet-Related Socio-Economic Issues, pp.354–367.

Land, S.M., Greene, B.A., 2000. Project-based learning

with the world wide web: A qualitative study of

resource integration. Educational Technology

Research and Development, 48(1), pp.45-66.

Larusson, J., Alterman, R., 2009. Wikis to support the

“collaborative” part of collaborative learning.

International Journal of Computer-Supported

Collaborative Learning, 4(4), pp.371-402.

Ma, W., Yuen, A., 2008. A Qualitative Analysis on

Collaborative Learning Experience of Student

Journalists Using Wiki. In Hybrid Learning and

Education. pp. 103-114.

Pusey, P., Meiselwitz, G., 2009. Heuristics for

Implementation of Wiki Technology in Higher

Education Learning. In Online Communities and

Social Computing. pp. 507-514.

Sbarski, P. et al., 2008. Visualizing Argument Structure.

In Advances in Visual Computing. pp. 129-138.

Scheuer, O. et al., 2010. Computer-supported

argumentation: A review of the state of the art.

International Journal of Computer-Supported

Collaborative Learning, 5(1), pp.43-102.

Suthers, D.D. et al., 2008. Beyond threaded discussion:

Representational guidance in asynchronous

collaborative learning environments. Computers &

Education, 50(4), pp.1103–1127.

CSEDU 2011 - 3rd International Conference on Computer Supported Education

364