=Paper=
{{Paper
|id=Vol-1736/paper3
|storemode=property
|title=Visualizing Online (Social) Learning Processes - Designing a Dashboard to Support Reflection
|pdfUrl=https://ceur-ws.org/Vol-1736/paper3.pdf
|volume=Vol-1736
|authors=Darya Hayit,Tobias Hölterhof,Martin Rehm,Oskar Carl,Michael Kerres
|dblpUrl=https://dblp.org/rec/conf/ectel/HayitHRCK16
}}
==Visualizing Online (Social) Learning Processes - Designing a Dashboard to Support Reflection==
https://ceur-ws.org/Vol-1736/paper3.pdf
Visualizing online (social) learning processes –
Designing a Dashboard to support reflection
Darya Hayit, Tobias Hölterhof, Martin Rehm, Oskar Carl, Michael Kerres
University of Duisburg-Essen, Essen, Germany
darya.hayit@uni-duisburg-essen.de
Abstract. Learning analytics, as a means to visualize learning, has been re-
peatedly suggested to enhance learners’ and teachers’ self-reflection in online
learning processes. Departing from this notion, we propose a combination of
this visual approach to learning analytics with the concept of social presence,
thereby acknowledging social aspects of online learning processes that are of-
ten overlooked. More specifically, we present the considerations and design
of a dedicated dashboard that supports self-reflection by visualizing (social)
online learning processes. The approach is based on our belief that visualiz-
ing learning by itself does not automatically lead to self-reflection and
awareness among students and teachers. Instead, organizers and instructors of
learning activities need to be conscious about the social aspects of learning.
Keywords: Dashboard, reflection, social learning, online
learning, awareness, visualization, learning analytics, social presence
1 Types of Learning Analytics – A German Perspective
The current discussion on learning analytics is based on two main approaches: The first ap-
proach focuses on the possibility of using learning analytics as a means to visualize learn-
ing, create awareness and stimulate self- reflection [1, 2]. The second approach centres
around the idea of stimulating learning through programmed instruction (e.g. adaptive sys-
tems) by guiding learners through the learning process [3, 4]. Hence, it can be stated that
the role of technology within these two approaches is different. While the latter approach
assigns technology a more active role – intervening and guiding the learning process – the
prior approach focuses more on technology as a formative tool – visualizing the learning
activities in order to stimulate reflection and awareness of the underlying learning processes
[5].
When considering the German discussion about this topic, the technology-driven ap-
proach is widely criticised and often even rejected as a methodological approach to inform
students and teachers. Among the most commonly mentioned reasons for this position are
concerns about privacy issues and, more importantly, doubts about employing an automated
system to influence and intervene into the learning process of individuals. Consequently,
the visual approach to learning analytics appears to be a more promising point of departure
35
Visualizing Online (Social) Learning Processes - ARTEL16
when considering the implementation of such systems in a German context.
Dashboards are a frequently used and investigated tool in learning management systems
to visualize learning activities. They consist of dedicated pages or areas within the system
mirroring the personal learning process and thereby contributing to the perception and re-
flection of underlying learning processes. [6] Moreover, departing from the community of
inquiry framework (CoI), it is possible to make other participants in a learning environment
visible at all three levels: cognitive presence, social presence and teaching presence [7].
However, we believe that visualizing learning by itself does not automatically lead to
self-reflection and awareness among students and teachers. Instead, drawing on recent con-
cepts of online learning, like the CoI or the 3C model, the social dimension of learning
might need to be emphasized more strongly. Accordingly, we argue that organizers and in-
structors of learning activities need to be conscious about the social aspects of learning.
Many systems seem to focus on the interaction between the learner and the technology (e.g.
often the Learning Management System wherein the learning activity is hosted and provid-
ed). We propose to extend this approach and to also incorporate the social aspects and inter-
actions between learners in the visualization of learning, thereby providing a more complete
representation of the underlying learning processes. It is the aim to contribute to the person-
alization of a learning environment [8].
Both concepts, the CoI as well as the 3C model, distinguish between three aspects of
learning. While the CoI model focuses on three kinds of presence, namely teaching pres-
ence (e.g. direct instructions), social presence (e.g. emotional expressions, group cohesion,
open communication) and cognitive presence (e.g. triggering events and exploration) [7],
the 3C model has a somewhat different focus. Following this model, an online course con-
sists of content (e.g. various kinds of presenting information), construction (e.g. learning
tasks) and communication (e.g. video conferences, chats and/or forum discussions) [9]. To
some extent, those models share the same perspective on online learning: beside emphasiz-
ing the social dimension of learning they mention its cognitive component, as well as the
need for instruction. Accordingly, dashboards to mirror a social learning processes consists
of three components containing visualizations of those dimensions.
Previous research explored a link between social interaction in learning management
systems (LMS) and learners’ social presence. Among others, Hölterhof and Rehm (2016)
combined the results of social network analysis and social presence, in order to determine
learners’ position within a communication network and relating this to their (social) experi-
ences within the LMS in question. More specifically, the authors were able to unveil differ-
ent dimensions of social presence, especially pointing towards positive as well as negative
social emotions. While research often focuses on positive emotions, both sides of socio-
emotional awareness of other learners are important for a technology enhanced social learn-
ing process, especially if learning is considered as an inquiry process. [10]
Following this approach of not assessing learners experience of social presence but to visu-
alize the social heterogeneity of learning as a group inquiry process, the advances of learn-
ing analytics turns towards transparency in providing these type of results to all relevant ac-
tors in the learning process (e.g. learners and teachers).
In order to take into consideration both the course structure on the one hand and social pro-
cesses within the structures on the other hand, we develop a dashboard based on the afore
mentioned 3C-model.
36
Visualizing Online (Social) Learning Processes - ARTEL16
2 Designing a Dashboard to Visualize (Social) Learning Pro-
cesses
Departing from the aforementioned considerations and stipulations, we identified a high po-
tential for a technology based solution to support and raise awareness in the context of ena-
bling social presence (experiences). Consequently, we are in the process of designing a
learning analytics dashboard, which is envisioned to become a feedback instrument, sup-
porting learners in self-reflecting their learning progresses. The dashboard is integrated in a
social learning management system based on the content management system (CMS) Dru-
pal®, which is enriched by numerous features to enable communication and collaboration
between learners and teachers [11, 12]. The system is further extended by a range of cus-
tomized modules that visualize the underlying social and cognitive learning processes.
Drawing on the 3C-model of online learning, digital learning contains of three different
types of structural elements: content, construction and communication. The dashboard de-
picts all three elements of the model and is based on a selection of different applicable vari-
ables. The selected variables per category arise from the various affordances that the LMS
offers. The content component visualizes the usage of learning materials available to learn-
ers, like text documents, interactive trainings or videos. Especially clicks on learning mate-
rials are considered to represent its usage. The constructive component mirrors learners’
behavior in relation to the learning assignments. Visualizations within this component pre-
sent the number of learning tasks per course unit, the number of submissions per task and
the number of tries per person in order to solve a learning tasks. The communication com-
ponent can be considered as rudimentary perspectives on social structures similar to what
social network analysis investigates. They offer interpersonal communication, including
number of posts in a discussion forum, comments per post and a distribution of posts per
role (teacher and learner). Table 1 gives an overview of other variables, which will be pre-
sented within the dashboard.
In order to enable a possible transfer of the dashboard into other CMS and LMS (e.g. Moo-
dle), the chosen variables and database structure have been constructed to enable this in-
teroperability.
37
Visualizing Online (Social) Learning Processes - ARTEL16
“Content” variables “Construction” variables “Communication”
variables
Number of discussions/
Number of learning materials Number of learning tasks
posts
Number of submissions per Number of comments per
Usage of learning materials
learning task and person post (in average)
Table: Number of tries per
Distribution of posts per
Proportion of usage person in order to solve a
role (teacher / learner)
learning task
Number of persons per Percentage of posts and
number of completed comments per role (teacher/
learning tasks learner)
Number of persons who only Wordcloud with frequent
needs one try to pass the task words
Table 1. Dashboard variables
The dashboard will be piloted in the context of two online master study programs at a Ger-
man University, which are designed as in a blended learning course format.
The programs focus on online-learning- periods, which last at least nine weeks and up to
twelve weeks in which three weeks form a unity. During this time, participants communi-
cate with each other and engage into learning activities within the applicable LMS. Ulti-
mately, the goal of this instrument is to stimulate course (activity) by enhancing transparen-
cy of (social) learning activities at different points in time. After each three-week course
unit, students and teachers will be able to voluntarily access the current visualization of
what activities took place. This in turn creates an opportunity for all participating actors
(e.g. learners and teachers) to self-reflect about their learning behavior. It also relates effec-
tive data to previous points in time as well as previous courses. A visual representation of
an initial wireframe is provided in Figure 1 below.
38
Visualizing Online (Social) Learning Processes - ARTEL16
Fig. 1. Initial Wireframe of Dashboard
References
1. Duval E (2010) Attention Please! Learning analytics for visualization and recommen-
dation. In: ACM (ed) Proceedings of LAK11: 1st international conference on learning
analytics and knowledge, New York, 2011.
2. Leony A, Pardo A, de la Fuente L, Sanchez C, Delgado C (2012) GLASS: A learning
analytics visualization tool. In: ACM (ed) Proceeding LAK ‘13 proceedings of the
third international conference on learning analytics and knowledge, New York, 2012.
3. Chen W, Persen R (2009) A recommender system for collaborative knowledge. In: IOS
Press (ed) Proceedings of the 2009 conference on artificial intelligence in education,
Amsterdam, 2009.
4. Özyurt Ö, Özyurt H, Baki A, Güven B (2013) Integration into mathematics classrooms
of an adaptive and intelligent individualized e-learning environment: Implementation
and evaluation of UZWEBMAT. Computers in Human Behaviour 29(3): 726-738.
39
Visualizing Online (Social) Learning Processes - ARTEL16
5. Ruipérez-Valientea JA, Muñoz-Merinoa PJ, Leonya D, Delgado Kloosa C (2015)
ALAS-KA: A learning analytics extension for better understanding the learning pro-
cess in the Khan Academy platform. Computers in Human Behaviour 47: 139-148.
6. Verpoorten D, Westera W, & Specht M (2011) A first approach to “Learning Dash-
boards” in formal learning contexts. Paper presented at the ADVTEL workshop (1st
International Workshop on Enhancing Learning with Ambient Displays and Visualiza-
tion Techniques) at EC-TEL 2011, Palermo, 20 Sept 2011.
7. Garrison DR, Anderson T, Archer W (2000) Critical Inquiry in a Text-Based Environ-
ment: Computer Conferencing in Higher Education. The Internet and Higher Educa-
tion, 2(2-3), 87-105.
8. Verpoorten D, Glahn C, Kravcik M, Ternier S, Specht M (2009) Personalisation of
Learning in Virtual Learning Environments. In Cress U, Dimitrova V, Specht M (Eds.)
Lecture Notes in Computer Sciences 5794, Learning in the Synergy of Multiple Disci-
plines, 52-66. Berlin, Germany, Springer-Verlag.
9. Kerres, M, de Witt, C (2003) A didactical framework for the design of blended learning
arrangements. Journal of Educational Media, 28, 101-114.
10. Hölterhof T, Rehm M (2016): Learning in Online Social Networks: Exploring Structur-
al and Emotional Aspects of Social Presence and Sense of Community. Paper presented
at AERA 2016, Washington D.C., 11 Apr 2016.
11. Hölterhof T, Heinen R (2014) A Concept to Bridge Personal Learning Environments:
Including a Generic Bookmarking Tool into a Social Learning Management Systems.
Journal of Literacy and Technology, 15(2), 111–135.
12. Hölterhof T, Nattland A, Kerres M (2012) Drupal as a Social Hub for Personal Learn-
ing. In: PLE Conference Proceedings. Available via
http://revistas.ua.pt/index.php/ple/article/view/1453. Accessed 29 Jun 2016.
40