Building a Tool for Analyzing Interactions in a Virtual Learning

Environment

Leticia Rocha Machado, Magali Longhi and Patricia Behar

Federal University of Rio Grande do Sul, Paulo Gama Avenue,

110 - Building 12105-3 floor room 401 Cep: 90040-060, Porto Alegre (RS), Brazil

Keywords: Interactions, Virtual Learning Environment, Social Map.

Abstract: This article presents the development of the framework Social Map, implemented in a virtual learning

environment (VLE). The tool aims at mapping the interactions of the participants of a course in the form of

graphs or sociograms. The methodology used was descriptive, theoretical, and practical and the graphs

generated from the interactions of two undergraduate classes and one continuing education course were

analyzed. Data from the Social Map, and a punctual analysis of the interactions and relationships, enabled

the teachers to rethink educational activities in VLEs. Based on the indicators obtained from the map, the

aim is to build more consistent teaching practices for activities in the virtual, especially concerning the

social aspects. It is also a way to highlight peculiarities hitherto little-discussed regarding distance

education.

1 INTRODUCTION

This article aims to present the development of the

framework Social Map, which shows, in graphs,

information concerning student interactions obtained

in a virtual learning environment (VLE) know as

ROODA.

The research has an interdisciplinary character as

it brings together the fields of Sociology,

Information Technology, and Education. Piaget’s

ideas (1973, 2005) laid the foundations for the study

of social interactions in Education and underlie the

VLE used in this research. The investigation also

relies on Behar’s studies (2009, 2013) to understand

the process of teaching and learning in Distance

Education (DE) and the construction of teaching

strategies applied in the virtual. Regarding

sociology, the research draws on Moreno’s theory

(1954) when dealing with sociometric analysis,

known as the mapping process of interactions and

presented as sociograms.

Data processing contributes to the investigation

with regard to the concepts addressed in graph

theory (Wasserman, Faust, 1994) and the study of

techniques for computational implementation.

In this sense, studies in different areas of

knowledge subsidize the proposal of a

computational solution that tries to ensure,

particularly in the field of distance education,

teaching and learning that allows the teacher to get

sociometric information about students, for it

considers that such information should be taken into

account in the process of distance teaching and

learning

The research question of this study refers to the

way the teacher can view the student(s) VLE social

behavior(s). It proposes the building of a feature that

maps and presents, in the form of graph, the

interactions established by a particular student or a

group.

Thus, the investigation followed two lines of

work: one focused on the social aspects of

theoretical character and another on the planning,

implementation and validation of the framework

Social Map in the VLE. The project dedicated to the

development of the Social Map feature began in

2013 with the SocialAffective Research Group

(http://www.ufrgs.br/gpsocioafeto/) at the Federal

University of Rio Grande do Sul (UFRGS), Brazil..

This feature collects data from interactions held in

the VLE communications features, and presents as

sociograms relations between the actors of the

discipline and/or course. Sociograms are viewed

from the GraphViz tool, free software to represent

information in graphs. These graphs are presented to

Machado, L., Longhi, M. and Behar, P..

Building a Tool for Analyzing Interactions in a Virtual Learning Environment.

In Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2015) - Volume 3: KMIS, pages 287-291

ISBN: 978-989-758-158-8

287

the teacher in PDF or image formats that can later be

saved on the computer.

As for the theory, the social characteristics which

could be the basis for the development of the

research and the improvement of the Social Map

feature were discussed. In order to do so, Moreno's

ideas (1972) and the literature referring to the

distance mode (Behar, 2013; 2009) were used as

resources to list the features.

The evaluation and the improvement of the

digital resource considered the observation and

analysis of the social mapping resulting from the

tool application in two undergraduate and one

continuing education course.

To understand the trajectory used in the

construction of the Social Map tool, this paper

presents, in section 2, the theoretical framework.

Section 3 presents the methodology used in the

investigation. Section 4 includes the process of

implementation, validation and the results of the

Social Map. Finally, in section 5, the final

considerations are made

2 THEORETICAL

BACKGROUND

In distance education, the main protagonist of the

teaching and learning process is the student (Behar,

2009). For Piaget (2005), mutual respect, autonomy,

and cooperation are characteristics of a socially and

morally developed subject. These elements are built

along the stages of life promoted by interactions

with others, with objects, and with the environment.

Etymologically the term interaction (inter +

action) includes the concepts of reciprocity, in which

at least two elements (they do not need to be of the

same nature) are involved; and contact, an encounter

that causes changes in the participating elements.

For Piaget (1973) it is in the interactions that the

subjects will build knowledge, for “social life is one

of the essential factors in the formation and growth

of knowledge” (Piaget, 1973, p.17).

Interaction may constitute intrapersonal or

interpersonal level. The practice of conversation

(and social relations formed from it) supported by

digital technologies can be mistaken for a message

flow where there is not necessarily a conversation or

social interaction (often referred to as interactivity).

It is understood that interactions in technological

spaces are based on a dialogue that modifies the

subject, the other, their messages and their

interrelations (Longhi, 2011).

According to Moreno (1972), sociometry is the

mathematical study of the psycho-sociological

properties of the population by putting into practice

an experimental technique based on quantitative

methods. According to the author, sociometry is a

strategy for understanding the structure of a group.

One of the techniques of Sociometry is the

application of sociometric tests that enable the

visualization of the similarities and differences

between individuals that make up a group.

Investigating interpersonal relationships in groups,

Moreno (1972) found that there are two main ones:

the relationships of attraction and repulsion. From

these are drawn many others. For example, the

investigated subjects express: a) their choices

regarding the colleagues that would like (or not)

help to perform a certain activity; b) the colleagues

that do better (or worse) at playing a certain role in

the group. Based on their choices, the mutual

relations of the subjects investigated are identified

and presented as a graph, known as sociogram,

which, according to Moreno, reveals even the

“invisible”.

The sociogram shows the position occupied by

the individual in the group and the core of

relationships that form around him or her. This core

of relations is the smallest social structure that

Moreno defines as social atom (Moreno, 1972).

While certain social atoms are limited to individuals

who participate in it, some of these individuals can

relate to parts of other social atoms, and so on,

forming complex chains of interrelations, which the

author calls sociometric networks.

Then, through a sociogram, which is a structured

visual representation of a network, the social

position of each participant in a learning community

and the relationship with the rest of the group can be

seen. Through the choices made, it is possible to

determine who in the social atom is the most

privileged and those who exert reciprocity; which

individuals are rejected for not fulfilling reciprocity;

and which are isolated for not showing their

preferences.

Sociograms are graphical representations, in the

form of a network, of the relationships in a group of

individuals. More than one method of presentation,

sociograms constitute a method of exploration, as

they enable the identification of sociometric facts

and the structural analysis of a group.

Moreno (1972) defined a set of symbols

(geometric shapes, such as circles, triangles with

single or duplicate edges, and straight lines with

continuous or broken lines, presenting or not arrows

in red or black), manually drawn representing the

KMIS 2015 - 7th International Conference on Knowledge Management and Information Sharing

288

genre of the subjects investigated, the role in the

group, the attractions and the repulses, indifference

and unilateral or bilateral relations.

In the 1960s, Moreno’s sociogram incorporated

formalisms of graph theory which gave it a

mathematical rigor (Wasserman and Faust, 1994)

and it started to be described by computer

algorithms with graphical display on various

devices. Currently, sociograms are recognized as

social networking diagrams.

The network (or graph) consists of a finite set of

nodes of the actors (individuals, groups, or

organizations) and edges (or arcs) indicating the

connections among them. When reading a graph, the

main focus of analysis is the pattern of the

connections regarding the distance and physical

location of the nodes. Thus, the analysis of

sociograms enables the verification of how

individuals relate, the choices they make and the

reciprocity among them. The distance and physical

position of the participants provide information

about the leaders, the isolated ones, and the

subgroups (or social atoms). In this study, these

definitions were termed as social characteristics. The

definition of the social features demanded the need

to know the peculiarities of each VLE interaction

feature. The aim was to identify how each clue could

support the mapping of social relations. This way,

the primary features outlined are: Collaboration,

Popularity, Isolation, Mediation, Subgroups, Social

Detachment, and Indifference.

3 METHODOLOGY

The investigation starts from the problem of how to

map the social interactions in a VLE so as to display

them in the form of graphs (sociograms). The

research is characterized as descriptive, theoretical,

and practical as it dedicates to the (re)construction of

ideas and improvement of theoretical principles,

mainly those related to studies of sociometric

aspects (Moreno, 1972).

Thus, in order to meet the proposed objectives,

the study was developed in four stages, which were

carried out in a recursive sequence:

1) construction of the theoretical framework on the

themes: social aspects (Primo, 2008; Piaget,

1979) Sociometry (Moreno, 1972), educational

(Piaget, 1973; 2005; Sacerdote, Fernandes,

2013; Lima, Meirinhos, 2011) and Distance

Education (Behar, 2009);

2) planning and implementation of the Social Map

feature, whose interactions analysis module was

built in its own environment and the GraphViz

library (http://www.graphviz.org) was

incorporated into the display module;

3) validation of the Social Map feature in

undergraduate and continuing education courses

offered at UFRGS;

4) consolidation of the social characteristics based

on the theoretical framework and on the results

obtained in the application analysis. Such a step

may require improvements in the implemented

feature.

The Data collection instruments used were three:

a) participant observation; b) data collected through

the productions in the features of the VLE, and c)

graphs of the interactions generated by the

framework Social Map.

4 TRAJECTORY AND RESULTS

Data from this project are presented in the form of

graphs generated by the framework Social Map

where it was possible to analyze the social

interactions of students in two undergraduate

courses and one continuing education course in

addition to validating the developed resource.

The Social Map implemented in the ROODA

VLE is a feature that allows, from the interactions of

users in communication tools in the VLE, the

generation of sociograms where links, influences

and preferences present in the social group formed

by disciplines and courses can be identified. Such

feature is accessible in the moment, only to the

teacher of the course

The ROODA VLE allows various forms of

interaction among participants: through exchanges

of messages on forums, chat, or e-mail; comments

on the inclusion of materials in the library and

comments on certain forums messages, lessons and

activities. All these interactions are captured,

analyzed, and extracted in a text format file and then

sent to the Graphviz to generate the graph, which

enables us to see how interactions among

participants in a class, for example, can be seen.

In computational terms, the sociogram

construction process begins in a PHP class where the

constructor method established settings to generate

the social map. To this end, the following settings

must be supplied by the user:

• Analysis period: sets the length of time that

teachers want to see the interactions carried out;

• Colors of the participants: the teacher can specify

colors for each user profile (monitor/tutor,

teacher, and student). By default, the color

Building a Tool for Analyzing Interactions in a Virtual Learning Environment

289

orange represents teachers, lilac represents the

students, and gray represents the monitors (or

tutors);

• Interaction method: defines the type of

visualization of the interactions (bonds of a

participant with the class/group, all the students,

students and teachers, students and monitors, all

participants);

• Layout of the sociogram: depicts the format of

the sociogram. It has two types: graph showing

the interactions in a network (Figure 1) and

graph that shows the interactions in a circular

manner (Figure 2). The maps of Figures 1 and 2

were obtained from a distance course supported

by the ROODA VLE.

• Level of relevance: the teacher can assign

different levels of importance for each feature

analyzed. These important levels directly

influence the thickness of the edges that connect

the nodes. The levels range from “Not

applicable” to “extremely important”.

Figure 1: Layout of the Social Map in a network.

Figure 2: Layout of the Social Map in a circular shape.

For mapping the interactions, SQL queries to the

database where all the AVA information is stored in

physical tables are made. These SQL queries return

the data to the PHP class, which processes these data

and transforms them into information ready to be

displayed. Processing consists in counting the

interactions between users for features such as

forum, chat, contacts and comments on library

materials and webfolio. After this count, a text

description in DOT file is sent to GraphViz, which is

responsible for the generation of the graph,

according to the description.

The information analyzed always takes into

account the chosen class. Thus, the results in the

Social Map can be different for the same student,

when a different class is chosen.

Each tool analyzed receives an interaction

weight, which interferes with the thickness of the

lines connecting the nodes. In the case of a large

number of interactions, for example, the line width

is greater. This tells the teacher which students (and

monitors/tutors) had a greater number of interactions

during the selected period.

To improve the Social Map feature in the VLE,

the analysis of the interactions of two undergraduate

courses at UFRGS were used. One was offered in

the classroom mode (Figure 3) and the second in the

distance mode (Figure 4).

Figure 3: Classroom course.

Figure 4: Distance Course.

KMIS 2015 - 7th International Conference on Knowledge Management and Information Sharing

290

In Figures 3 and 4, nodes in orange represent the

students, the teachers are lilac and the

monitors/tutors are gray, for each discipline and/or

course.

In the Sociogram in Figure 3, taken from the

discipline taught in a classroom, the interactions are

directed to the classroom teacher, showing a lack of

interaction among students. In Figure 4, although

with a lower number of students, it is possible to

notice a larger number of interactions among

students themselves and between students and

teachers.

It should be noted that the evaluation of these

sociograms did not take into account the pedagogical

strategies undertaken by the course teachers, only

the graphical potential of the Social Map as a source

for teachers to examine and perhaps rethink their

courses. The extracted visual data suggest several

questions about the social interactions that take place

in VLEs.

Thus, the sociograms generated point to the need

for further investigations that include, in addition to

quantitative data, qualitative data in terms of social

relations in order to automatically display

possibilities for educational activities for teachers of

distance education. Therefore, new perspectives

suggest some improvements, as it will be presented.

5 FINAL CONSIDERATIONS

The present study showed the planning, the

development and the implementation of the Social

Map framework for the mapping of social

interactions. From the extracted data, the teacher can

analyze the possible social characteristics that are

present in interactions from the VLE communication

features.

For future research, the aim is to discuss and

develop, alongside the technological process,

pedagogical strategies that might help the teacher in

the pedagogical use of the Social Map. To improve

this framework, the following actions are being

taken: (1) building dynamic graphs in order to

improve the visualization of results; (2) inclusion of

new social features (or modifying existing ones) to

better understand the relationships that are formed in

a VLE; (3) studies to make available the maps for

the students; and (4) in computational terms,

performance and application studies in other

distance learning platforms;

The main contribution of this paper is to present

the Distance Education teacher information on some

social features that can be recognized in a virtual

learning environment. From the graphical view of

these characteristics, the pedagogical practices may

be redirected in order to individualize assistance to

students. The expansion of communication between

teacher and students is also envisaged as a

contribution.

REFERENCES

Almeida, L. 2011. Analisando Participação em Turmas de

Ensino a Distância Usando Análise de Redes Sociais,

Simpósios Nacionais de Tecnologia e Sociedade.

Bakharia, A., Dawson S. 2011. SNAPP: A Bird’s-Eye

View of Temporal Participant Interaction, Learning

Analytics and Knowledge.

Behar, P.A. 2009. Modelos Pedagógicos em Educação a

Distância, Porto Alegre, Artmed.

Behar, P.A. 2013. Competências na EAD, Porto Alegre,

Grupo A.

Gansner, E., Koutsofios, E., North, S. 2006. Drawing

graphs with GraphViz. Disponível em:

http://www.graphviz.org/Documentation/dotguide.pdf.

Lima, L., Meirinhos, M. 2011. Interacções em fóruns de

discussão com alunos do ensino secundário: uma

análise sociométrica. In: VII Conf. Internacional de

TIC na Educação.

Longhi, M. T.: 2011. Mapeamento de aspectos afetivos em

um ambiente virtual de aprendizagem, Doctoral thesis,

PPGIE/UFRGS, Porto Alegre, RS.

Moreno, J.L. 1972. Fundamentos de la Sociometria,

Buenos Aires, Paidós.

Piaget, J. 1973. Estudos Sociológicos, Rio de Janeiro,

Forense.

Piaget, J. 2005. Inteligencia y afectividad, Buenos Aires,

Aique Grupo Editor.

Primo A. (2008. Interação mediada por computador:

comunicação, cibercultura, cognição, Porto Alegre,

Sulina.

Rabbany, R., Takaffoli, M., Zaïane, 2011. Analyzing

participation of students in online courses using social

network analysis techniques, Proc. of Educational

Data Mining.

Romani, L. 2000. InterMap: Ferramenta para

Visualização da Interação em Ambientes de Educação

a Distância na Web, Masters dissertation,

IC/Unicamp.

Sacerdote, H.C.S., Fernandes, J.H.C. 2013. Investigando

as interações em um ambiente virtual de aprendizagem

por meio da análise de redes sociais, Inf. e Doc., v. 4,

n. 1.

Wasserman, S., Faust, K. 1994. Social network analysis:

methods and applications, Cambridge, Cambridge

University Press.

Willging, P.A. 2008. Técnicas para el análisis y

visualización de interacciones en ambientes virtuales,

REDES- Revista hispana para el análisis de redes

sociales, v.14,n.6.

Building a Tool for Analyzing Interactions in a Virtual Learning Environment

291