zmWmQs

Evaluation and field trials of MOOC Platforms in the Spanish-speaking community
Ignacio Despujol, Carlos Turro, Jaime Busquets and Vicent Botti
videos have to be stored in YouTube and code is hosted in
Google app engine. These characteristics can be viewed
both as advantages and disadvantages. In the overall run
we have found the platform very solid, simple to administer
and scale and we liked the ability to add features by
modifying simple python scripts. It is also very cheap, with
bills of less than $1 a month for courses with 1800 students
enrolled.
On the other hand it completely lacked a teacher interface,
meaning teachers had to export their course information
in .csv and create their tests in a text file using a
pseudo code language, so it wasn’t suitable for non-technical
users. We took a workaround through the use of a
neutral production format, but this is not a solution for the
long run.
A lot of the shortfalls we found have been addressed in
later versions of CB, and the latest version is prepared for
several courses, which lets internet users see the units of
a course without being enrolled, has a teacher interface
(although it is still too tech oriented) and has incorporated
features that we don’t have, like peer to peer activities
and the recording of all the in lesson activities taken by a
student.
However it stills lacks a course information page and,
more importantly, the tools for communication with students
are very limited. It doesn’t have the possibility to
send emails to students, and the lack of integration with
Google groups and their limited capabilities as a forum
tool is a major drawback.
MiriadaX analysis
MiriadaX is a platform developed by Telefonica using Liferay.
It is a platform specifically designed for MOOCs.
When we used it, it allowed you to use videos from You-
Tube or upload them, incorporate html only pages and
activities (a, b, c quizzes or P2P) and it had its forums and
teacher interface.
Its major drawback at the time we used it (first half of
2013) was the very low usability for teachers. The options
for activities were limited and the peer to peer activities
were very problematic. It was impossible to work with
mathematical formulas without making a tedious image
cut and paste. There were some screens to follow student
progression, but it was impossible to download student
data.
So, in summary, we think that the version of MiriadaX
we used should be considered a beta edition, with many
things to solve (even as some of them were solved during
the courses).
Results
To gather demographic data in the upvx.es courses we included
an initial form that the students have to fill when
they enroll. We asked them: their name, gender, birth year,
education level, country of origin, city of residence and
how did they know about the course, letting them choose
which questions they wanted to answer. In MiriadaX we
didn’t have access to the initial survey made by the platform,
so we included a starter unit in the courses with our
own survey that included the same questions plus how
much time did they have to dedicate to the course and
how good was their previous knowledge on the subject. In
the last survey we included the “Which is your motivation
to take the course” question.
To gather satisfaction data from the students we prepared
an online survey with questions to evaluate the
system, the courses and the platform. In the first edition
on upvx.es (the GCB implementation) the survey was only
sent to the people that earned a credential as a prerequisite
to download it. In the MiriadaX edition the survey was
sent at the end of the courses to all students by mail, as
we didn’t have control on the issuing of credentials. In the
second upvx.es edition the survey was sent at the end of
the courses to all students by mail and was a prerequisite
to download the credential (if the student had passed the
course).
We can see that in the MiriadaX edition 75% of the students
took the first exam and in the GCB editions only ¼ of
them did (is the only way we have to know if they started
the course or not). We think that the reason was the lack
of email capabilities of GCB that didn’t let us reach a lot of
enrolled students. On the other hand the % of credential
issued is similar in the three editions, so the retention rate
of GCB is bigger.
Student demography
Students enroll mostly from Spain (66%), next from Colombia
(9%), Mexico (6%) and Peru (5%). When we consider
the ones that earn a credential the % that comes from
Spain grows to a number between 72% and 79% depend-
EDITION COURSES Initial answers Final answers Enrolled First exam Credentials Cred/First Exam
GCB 1 2 940 187 1171 295-25,2% 160-13,7% 54,2%
MIRIADAX 14 46341 8095 76459 57098-74,7% 11805-15,4% 20,7%
GCB 2 8 6700 850 5971 1416-23,71% 667-11,2% 47,1%
Experience Track |211