Similar a VII Jornadas eMadrid "Education in exponential times". Mesa redonda eMadrid. Analytics in Open Education. Edmundo Tovar, UPM. 04/07/2017.(20)
VII Jornadas eMadrid "Education in exponential times". Mesa redonda eMadrid. Analytics in Open Education. Edmundo Tovar, UPM. 04/07/2017.
1. VII Jornadas eMadrid
Round Table: "Learning Analytics"
Analytics in Open Education
Madrid, 4 de julio de 2017
Edmundo Tovar Caro
etovar@fi.upm.es
Technical University of Madrid
CC BY Edmundo Tovar
2. Learning Analytics (LA)
— “The measurement, collection, analysis and reporting of
data about learners and their contexts, for purposes of
understanding and optimizing learning and the
environment in which occurs”
Long, P., & Siemens, G. (2011). Penetrating the
fog: Analytics in learning and education.
Educause Review, 46(5
4. A way of carrying out education, often using digital
technologies. Its aim is to widen access and participation to
everyone by removing barriers and making learning accessible,
abundant, and customisable for all. It offers multiple ways of
teaching and learning, building and sharing knowledge. It also
provides a variety of access routes to formal and non-formal
education, and connects the two
Open Education is seen as
Source: JRC IPTS Report: Opening up Education: a support framework for higher education institutions.
http://publications.jrc.ec.europa.eu/repository/handle/JRC101436
Open Education
CC BY Jonatan Castaño Muñoz
Jornada red.es-UC3 sobre formación digital con tecnología abierta
5. Profiles of MOOC participants
Problem analysis
ü MOOC format is characterized by the great diversity of enrolled
people
ü There is a lack of knowledge of participants’ background
ü It is needed to provide a real picture of the heterogeneity of MOOC
participantsº
7. New measurements needed
— Different audiences than traditional HE degrees
— Different intentions and motivations
Source: online course report State of the MOOC 2016: A
Year of Massive Landscape Change For Massive Open Online
Courses
CC BY Jonatan Castaño Muñoz
Jornada red.es-UC3 sobre formación digital con tecnología abiert
8. For whom?
— Main actors in this scenario: MOOCs providers and EC
— Cross provider issues to learn more about the population of
MOOC learners and to develop policy guidelines
— Benchmarking reports will provide MOOC providers insigths
about their MOOCs compared with other MOOCs
10. MOOCKnowledge project
Initiative launched by the Institute of Prospective Technological Studies (IPTS).
Executed by the Open University of the Netherlands (OUNL) in cooperation with the Open
University of Catalonia (UOC) and Technical University of Madrid (UPM)
MOOCKnowledge pursues to assess the current perspective of
learners as participants of European MOOCs: when, why and
how the participants enrolled in one or more MOOCs take
advantage of them
Goal: building a [large scale] and cross-provider data collection
based on the experiences reported by the participants of the
MOOCs offered by providers that collaborate with the project
with the intention of a more deepen understanding of issues
related to MOOC format
Online multilingual
survey:
• Pre-questionnaire
• Post-questionnaire
• Follow-up questionnaire
Pre-questionnaire:
• attibutes:
demographics,
linguistics, motivation,
intentions, LLL, MOOC
experience, prior
satisfaction of
participants with MOOC
format
14. 74 participants with high languages
proficiency, reasonable motivation before their
enrollment in a new course, and without
previous experience in MOOC format
214
211
212
Preliminary findings: extended profiles
104 participants with high languages proficiency,
reasonable motivation before their enrollment in a
new course, and a high degree of previous
satisfaction in MOOC format
53 participants with high languages proficiency,
reasonable motivation before their enrollment in a
new course, and indifference regarding MOOC
format
16. Exploration of SNA to the
Moocknowledge data
— An initial study has been conducted on the potential of
social network analysis (SNA) and its metrics for:
— Analyzing data obtained from the answers the participants.
— Exploring useful scenarios.
— The data source used comes from 715 participants of
the MOOC on Test Anxiety Management.
17. Study case
— Scenario:
— Analysis of the behavior of students attending the different levels of
involvement to the MOOC who have registered
— Groups of interest: managers in open education, academic staff of
MOOCs,…
— Questions: C2B3_9, C2B3_10, C2B3_11, C2B3_12, C2B3_13 y C2B3_14
— Objective:
— To increase the size of the target cluster in next editions of the MOOC
— Phases:
— 1. Clustering of participants
— 2. Target cluster?
— 3. Identifying profile of students “bridge” between the target cluster
and other clusters
— 4. Promoting the participation of “bridge” students in the future
18. Study case.
1. Clustering of groups of participants
— The SNA analysis is based on the
similarity of participant profiles:
if two participants give the same
answer to a question, a
relationship exists between
them.
— The network consists of 316
nodes (participants) with 42117
relationships pointing.
— We obtained three clusters
represented with different
colors.
— Each cluster represents the
pattern of participants that have
the tendency to answer of the
same way.
HOW?
SNA metrics and filters:
• Weight (link)
Clustering: modularity metric method
After applying multiple filters to the network the number
of nodes is reduced to 306 and the number of edges to
23216.
19. Study case.
3. Identifying profile of students
“bridge”
— Profile of students contributing to the objective
— Second stage of tertiary education
— Employed for wages
— Woman
— Spanish
— Born in the 1960s
— Living together with partner
— Working in the Public administration and defense; compulsory
social security sector
HOW?
SNA metrics and filters:
Betweenness
21. Visualization Data
—Users should explore datasets even if the publisher of
the data does not provide any exploration or
visualization means.
—Applying information visualization techniques to the
data set helps users to explore large amounts of data
and interact with them.
22. Conclusions
ü Clustering is highlighted as a feasible technique for uncovering
participants’ profiles in MOOC format
ü Extended profiles as result of applying Data Mining or SNA
techniques
ü The identification of underlying relationships in the internal
structure of participants’ ofeatures might facilitate the making-
decisions process of policy-makers and educational-agents
regarding the target audience their institutions
Open Education, as well as MOOC format, could be positively
impacted by a more realistic understanding of people’s profiles