A Teaching Experience in a Theoretical Course based on Self-learning
Videos and Problem-solving Sessions
Marta Arias, Carles Creus, Adri
a Gasc
on and Guillem Godoy
Department of Llenguatges i Sistemes Inform
atics, Universitat Polit
ecnica de Catalunya, Jordi Girona 1-3, Barcelona, Spain
Self-learning, e-Learning, Videos, Teaching, Theory.
In this paper we describe a teaching experience applied to a theoretical course thought in a computer science
degree. The main feature of our experiment is the introduction of videos specifically designed for self-learning
as part of the learning process. Master classes are replaced by working sessions in which the involvement of
students gains prominence. The teacher explains almost nothing in class. Instead, most of the time is devoted
to the presentation of solutions to exercises assigned to students in advance. All presentations are done by
students, and the teacher only intervenes in order to complete explanations and correct mistakes.
The result of our experiment is promising from several perspectives. The exam results are better with the new
approach. The students learn to learn on their own and take better advantage of the time in class. The work
load is uniformly distributed along the course. The new approach also benefits the teacher since he/she spends
considerably less time preparing theory lectures, and gets continuous feedback to better follow the students
The videos are valuable in themselves and have been made publicly available. In fact, our students prefer
them to a master class. They can pause, rewind and replay the video, take a rest, and postpone the lecture if
necessary. Moreover, the interest for these videos goes beyond our university boundaries: according to the
visits’ place of origin and posted comments, they are being used by students from other countries.
Typically, lectures in mathematically founded sci-
ences include two different types of sessions: theory
and problems. The first type often consists of a set of
lectures in which the professor presents all the basics
on the subject, such as definitions and proofs of rel-
evant and useful results. This kind of content is well
established and does not lend itself to discussion.
The second part is devoted to the presentation, ei-
ther by the teacher or students, of solutions to prob-
lems previously given as homework. This second part
is often understood as a practice for the exam.
In our experience, theory lectures impose a
rhythm which is often difficult to follow by stu-
dents. Usually, theoretical concepts require an as-
similation time which exceeds the time allocated in
master classes, and it is different for every student.
Moreover, students tend to lose part of the teacher’s
explanation while taking notes. Their interaction with
the professor is limited to asking questions in case the
material is not well understood. In our opinion, the
teacher’s capabilities are not optimally exploited with
this kind of teaching approach, since active learning
approaches, and in particular problem-based learn-
ing, are especially well suited for scientific disciplines
(Handelsman et al., 2004). For this reason we believe
that active interaction in problem solving sessions
should be a central activity of the teaching process.
Educational theory suggests that there are better ways
to promote learning than the traditional master class
approach. The learning pyramid (see Figure 1), intro-
duced by the National Training Laboratories based on
Dale’s work (Dale, 1969), illustrates the average re-
tention of contents that corresponds to different teach-
ing methodologies, which is higher for audio-visual
material than for traditional lectures. In this vein, Tu-
tored Video Instruction (TVI, (Gibbons et al., 1977))
is a teaching method pioneered by Standford Univer-
sity. It consists in combining video-taped material
with teacher active tutoring. Our work falls into this
trend. The goal of this work is to combine video-
based material with problem-solving sessions where
the students actively present and discuss their solu-
Arias M., Creus C., Gascón A. and Godoy G..
LEARNING THEORY THROUGH VIDEOS - A Teaching Experience in a Theoretical Course based on Self-learning Videos and Problem-solving
DOI: 10.5220/0003333400930098
In Proceedings of the 3rd International Conference on Computer Supported Education (CSEDU-2011), pages 93-98
ISBN: 978-989-8425-50-8
2011 SCITEPRESS (Science and Technology Publications, Lda.)
Figure 1: The Learning Pyramid.
In recent years, there has been a great increase in
the amount of audiovisual teaching resources freely
available on the Internet (VideoLectures, 2010; Aca-
demicEarth, 2010; KhanAcademy, 2010). On the
one side, one can find commented slides and in-
formal explanations of a specific part of a sub-
ject (KhanAcademy, 2010). While this material is
well suited for solving particular questions when the
topic is already known, it is not enough for guiding
the learning process of a whole subject. On the other
side, one can find recorded lectures given by special-
ists on a certain topic (VideoLectures, 2010; Aca-
demicEarth, 2010). They are usually very interesting
since they allow us to attend a master class taught by
an important professor regardless of time and place.
The main disadvantage of these videos is that their
content has no added value with respect to the origi-
nal class, and has all of its defects. For instance, they
may be disordered or redundant at times, and contain
mistakes derived from the difficulty of the subject and
improvisation. These mistakes cannot be easily fixed
in the video.
For certain topics, teaching videos that are specif-
ically prepared for this purpose have several advan-
tages over regular master classes (Anderson et al.,
2001; Bauman and Jurgens, 2002; Day et al., 2005).
By careful preparation of content, image and audio,
one can produce a very succinct, well-ordered, ade-
quate and coherent material almost free of mistakes.
We have prepared videos of this kind for “Theory
of computation”, a course with a strong mathemat-
ical background of the computer science degree in
our University. These videos are intended as a self-
learning alternative to traditional master classes on
theory. Hence, they are self-contained, provide gen-
eral ideas, intuitions, precise definitions, proofs and
proposed exercises. The idea is to combine the pre-
cision and completeness of a textbook with the intu-
itions and dynamism of a good lecture.
Taking advantage of the videos, we have experi-
mented with the following teaching approach. Master
classes disappear and the teacher does not introduce
any material in class. Students learn theory on their
own by watching the videos, and are expected to solve
exercises assigned in advance. During class, students
explain their solutions on the blackboard as part of
the evaluation method. The teacher only intervenes in
order to complete explanations and correct mistakes.
One weekly session is devoted to study time, during
which students work on their own and the teacher is
present to help them, solving questions and giving in-
sights into the previously digested material. After this
continuous interaction with students, the teacher gets
a lot of feedback from each of them, making his/her
interventions more accurate and helping in the detec-
tion of common difficulties on the subject.
We applied our teaching approach to only one of
the groups of students of the course. The remain-
ing groups followed the traditional teaching method.
This allowed us to evaluate the impact of our method
versus the old one. The results are very promising:
students using the new approach were satisfied both
with the method and their performance on the sub-
ject. The professor that followed our method also ex-
pressed satisfaction with his new role in the course.
On the one hand, the method allowed him to influ-
ence the students’ learning process more positively.
On the other hand, the teacher felt free from the te-
dious part of the lectures thus improving his general
attitude towards teaching the subject.
2.1 The Course
We have applied our teaching method to “Theory of
computation”, a mandatory course taught during the
second semester of the third year of our computer
science degree. This course is of theoretical nature,
and is heavily based on mathematic principles. The
objectives of this course are mainly twofold: first,
to formally introduce the mathematical principles on
which computation is founded and, secondly, to teach
the students to follow and produce formal proofs and
solve problems in a rigorous manner. The contents
are hard, and students generally have difficulty with
the course. Moreover, it is mandatory, which alto-
gether results in a rather negative attitude towards the
subject. Despite the difficulty, the subject is interest-
ing, and the problems it presents can be very chal-
lenging and intriguing. This course has to compete
with other courses for the attention and dedication of
the students. These other courses tend assign more
applied work such as programming projects, which
are typically done in groups. Students spend a lot
of their time doing these projects, and concentrate on
CSEDU 2011 - 3rd International Conference on Computer Supported Education
this course just a week before the exam.
Students are divided into three groups, each taught
by a different professor. The lectures of two of the
groups are scheduled in the morning. The ones of the
third group are in the afternoon. We have applied our
new teaching methodology to the afternoon group but
allowing students to choose between the traditional
and the new evaluation method. The course has a
teaching load of three sessions of two hours each per
week (per group), the students are expected to devote
another 6 hours to self-study for an adequate assimi-
lation of the contents.
2.2 Comparison of Teaching
The main difference between the traditional and
new methodologies is the use of especially designed
videos to teach the theoretical basis of the course con-
tents. Another difference is in the evaluation method.
In the new methodology, participation in the class-
room is explicitly considered in the grading method.
In our system, grades are between 0 and 10 (10 being
outstanding). A nominal qualification is associated to
each mark range: 0 4.9 is failure, 5 6.9 is pass,
7 8.9 is good, 9 9.9 is excellent and 10 is with
honors. Students who fail (with a grade of less than
5) must take the course again. The results are usu-
ally not good. The percentage of students passing the
subject varies between 20% and 40%, resulting in a
high number of students taking this course 2 or even
3 times.
2.2.1 The Traditional Teaching Approach
In our university, “Theory of computation” is tradi-
tionally taught in a series of master classes. Two of
the three weekly sessions are devoted to the presen-
tation of theoretical contents by the professor. These
contents include definitions, fundamental results, and
formal proofs of the results. The third session consists
in solving problems where the concepts introduced in
the theory sessions have to be applied.
Lecturers usually encourage their students to par-
ticipate in class, with very little success. Partly, be-
cause participation is not explicitly considered in the
grading method. As a result, solutions are presented
by the teacher on the blackboard. For this reason,
problem solving lectures become in practice a mas-
ter class in which the students only copy the solutions
presented by the professor, with the hope of under-
standing this material while preparing for the final
Evaluation Method. The students take five tests of
30 minutes each throughout the course. These tests
are fairly easy, since in order to pass them it is suffi-
cient to carefully study some proofs given during the
theory lectures. Let C be the overall mark from all of
these tests, which is between 0 and 2.5. There is also
a final exam, whose mark, F, lies between 0 and 10.
The mark of a student in this subject is obtained by
the formula F + (1
) ·C. The formula may seem
strange at first. It is similar to a weighted average,
but note that the less F is, the more weight is given
to C. For example, if one gets a 10 in F (the best
grade possible), then the weight of C is 0. Thus, in
this case the value of C does not affect the final mark.
If one gets a 0 in F (the worst grade possible), then the
weight of C is 1. Thus, in this case the final mark is
just C. The formula benefits students who work dur-
ing the course, even if they perform badly on the final
exam. The following example illustrates this situa-
tion. If one gets a 4 in the final exam (F), the weight
of C is 0.6. In this case, the final grade is given by the
formula 4 + 0.6 ·C, which is between 4 and 5.5. So,
in this case, the student only passes if he/she obtains
a good grade on the tests (C).
2.2.2 The New Teaching Approach
We pretend to radically change the dynamics of lec-
tures, giving more emphasis to students’ involvement,
and making the professor’s role much more interac-
tive and fruitful.
Our new approach introduces both a different kind
of lecture and a new evaluation method. As men-
tioned above, students learn theory on their own by
watching the videos
, and the teacher does not intro-
duce any material in class. Instead, this time is used
by students to explain their solutions on the black-
board. Note that 4 hours of master classes are sub-
stituted in the new method by 4 hours of interactive
problem solving sessions. The third weekly session
is devoted to study time, during which the professor’s
only task is to solve and clarify questions. In other
words, the teacher is not grading or teaching, but is in
the class to help with the content. They can come with
their laptops, look at the videos again, ask questions
about them, and try to solve the assigned exercises
with possible hints given by the teacher. These ses-
sions create a positive atmosphere where the students
can more easily overcome their shyness and become
active members of the class.
Videos are posted on youtube and linked at the end of
the main subject’s page (www.lsi.upc.edu/˜ggodoy/tc.html).
LEARNING THEORY THROUGH VIDEOS - A Teaching Experience in a Theoretical Course based on Self-learning
Videos and Problem-solving Sessions
Evaluation Method. Given the nature of our insti-
tution and our students, participation does not come
easily. Therefore, we try to stimulate participation
by including it explicitly as part of the evaluation
method. To this end, the teacher scores presentations
of solutions by the students.
The new method has three parts: the results of
five tests taken during the course (C, the same ones
as in the traditional evaluation method), the result
of a final exam (F, also present in the traditional
method) and a new part, P, corresponding to students
presentations of solutions to problems on the black-
board. The formula used to compute the final grade is
min(10, 0.8·F + 0.6 ·C +P), where C ranges between
0 and 2.5, F ranges between 0 and 10, and P ranges
between 0 and 1.5. Note that with this methodology,
students can score a maximum of 11 points. This is to
encourage students to choose the new approach.
2.3 The Experiment
We have applied our new teaching methodology to the
afternoon group. The other two morning groups fol-
low the traditional approach. Our experiment consists
of comparing the performance of the students under
each of the methods applied. Note that the fact that
the new teaching approach is applied to the afternoon
group only, may be a drawback for the success of the
experiment: experience shows that the best students
are usually registered in the morning groups.
We believe that students should always have the
right to be evaluated with just a final exam, there-
fore we allowed students in the experimental group
(i.e., the afternoon group), to choose between the tra-
ditional and the new evaluation method. That is, they
can attend the lectures with the new method, but they
will be scored with the traditional approach. Notice
also that all students have access to the videos, includ-
ing the ones in the morning groups. Moreover, any
student can benefit from the problem solving sessions
and presentations by other students, even if he/she is
not required to do the presentations. Both these facts
may bias the result of the experiment against the new
method, since the newly created materials benefit all
the students. It is also true that students of the af-
ternoon group could attend the master classes in the
morning. However, an opinion poll conducted among
students (see Section 3.2 below) indicates that stu-
dents prefer the videos to the master classes.
3.1 Video Accesses
The videos are posted on youtube
and are public.
This site monitors the number and origin of video
sightings. We give a brief summary of this data. We
restrict ourselves to the videos in catalan and span-
ish, since the ones translated to english have not been
available long enough to give meaningful numbers.
The number of accesses varies widely depending
on the video. As of October 21st, 2010, the most
viewed video has 5419 accesses. The least viewed
one has 126 accesses. In the following table, we count
the number of videos for several ranges of accesses.
Table 1: Video accesses.
# accesses # videos
125-249 20
250-499 6
500-999 7
1000 4
Only 27 students chose the new approach. From
their comments, we know that each of them watched
each video twice on average. It is expected that some
of the other students watched the videos too. How-
ever, most of the accesses come from other countries,
mainly from Latin America. Since the topic is the-
oretical and very specific, we believe that the view-
ers are students from other computer science degrees.
This fact suggests that the videos are useful to learn
the subject in general and not only in the context of
our teaching approach. Not surprisingly, the most ac-
cessed videos are the ones introducing classic mate-
rial that is present in any course in theory of compu-
3.2 Students’ Opinion
We collected anonymous opinions from students
along the course. Some of the students were taking
the subject for the second time, so that they were able
to compare both methods. Here we summarize the
most frequent ones:
Being able to watch the video at one’s convenient
time is better than being forced to attend a lecture
at a fixed time, thus avoiding situations when one
is tired after several classes, has had a plentiful
lunch recently, or has a bad day.
The videos are posted on the channel
CSEDU 2011 - 3rd International Conference on Computer Supported Education
With the videos, one can pause the lecture at any
time, take a rest, rewind it, and see it again and
again if necessary.
With the old approach, once one gets lost in the
class, the rest of the time is useless. This problem
is solved with the new approach.
Presenting solutions of problems each week is a
good motivation to follow the course. The mate-
rial is better understood in this way.
The method forces one to work on the subject con-
tinuously during the course. The effort required is
constant during the course and not excessive. It
is deferential toward other subjects. The opposite
is not true in general. Other subjects have strong
peaks of work during the semester.
A big amount of exercises are solved in class.
This makes it more interesting, and seems a better
preparation for the final exam, which essentially
consists in solving exercises.
The first created videos are too slow. Neverthe-
less some students were able to accelerate them
making use of a computer application (the video
speaker is concerned about the quality and tone of
his/her voice as a consequence of such accelera-
3.3 Teachers’ Opinion
Here we summarize the impressions of the teacher
who applied the new method:
The fact that the students have to present their so-
lutions on the blackboard motivates them much
more than just delivering a paper with the solu-
tion. In fact, being exposed to their classmates
criticisms forces them to prepare well their re-
spective presentations.
The solutions presented by each student are very
interesting and rewarding for their classmates.
Even if such solutions contain mistakes, the dis-
cussion generated from them is very fruitful.
The professor is more accessible to the students
with the new approach. They frequently ask ques-
tions about how to find solutions and how to
present them.
After an intensive interaction with the students
during the course, the professor has more accu-
rate information of each student, such as his/her
knowledge on the subject, skills, strengths and
weaknesses. Thus, it is possible to take more as-
pects into account in addition to the ones evalu-
ated by a written exam in order to give a final mark
to each student.
1.5 points for the blackboard presentations are not
enough to motivate all students. Due to the high
workload imposed by other subjects, half of the
students did not solve the assigned exercises of
the last part of the course.
The preparation time and effort of the teacher
is lower. It is not necessary to prepare master
classes, which is hard and takes time. The con-
tinuous evaluation is done in the class, so it does
not impose additional correction workload with
respect to the old approach.
3.4 Students’ Results
An unavoidable point for the comparison between the
previous and the new teaching approach is the per-
formance of students on the subject. We will avoid
fine-grained comparisons since we consider that the
population is not large enough for a rigorous analysis.
Instead, we just present the result of our experience
letting the reader extract his/her own conclusions.
Table 2: Comparison of results.
Group 1 Group 2 Group 3
# students 46 0 50 0 19 27
# students passing 12 0 20 0 7 20
% students passing 26.1% - 40.0% - 36.8% 74.1%
Table 2 shows the number of students registered
for the course, the number of students who passed the
course, and the percentage of this two values, grouped
by the followed teaching approach. In the table, P
stands for previous teaching approach, and N stands
for new teaching approach. Recall that only students
of group 3 were allowed to choose between the two
approaches while all the members of groups 1 and
2 were evaluated and taught using the previous ap-
Finally, we would like to remark that applying the
previous evaluation method to the 20 students that
passed the subject being evaluated with the new ap-
proach, 18 of them pass the subject anyway. In other
words, 18 students pass the subject with the old eval-
uation method out of the 27 students to which the
new teaching approach was applied. This represents a
66.7% of students passing the subject. This strongly
suggests that the good performance of students that
followed the new approach is not due to a more re-
laxed evaluation method.
LEARNING THEORY THROUGH VIDEOS - A Teaching Experience in a Theoretical Course based on Self-learning
Videos and Problem-solving Sessions
In this Section we describe the video creation process.
Each video is at most 10 minutes long, since this was
the limit given by youtube for freely submitted videos.
This length actually works fine for us: a succinct and
well prepared video of 10 minutes may give as many
concepts as a class of 50 minutes, and it is more acces-
sible. This is because lectures are partly improvised,
the teacher can make mistakes, he/she can waste time
in complicated details which are difficult to keep in
mind, may explain related facts which are interesting
but not central, etc.
We have created 37 videos in total, mostly in
Spanish.They are being currently translated to En-
glish. In fact, most of them have been already trans-
lated to English.
Our video creation method has evolved until find-
ing a procedure with a good balance between qual-
ity and effort. The first couple of videos took about
a week each to make. While they have been a suc-
cess in the sense that they have over 5000 visits and
positive comments, they took too much time to make.
After that, we thought of more efficient ways to cre-
ate videos, and came up with the following process,
which we applied to all of the remaining videos.
1. First, we write a pdf document that contains the
image as well as the script of the video’s audio.
2. Secondly, we capture the audio. Any mistake is
immediately corrected by moving back the audio
recorder a few seconds.
3. Third, we mark a screen capture area and record
the image in this area while the audio is being
4. Finally, we combine audio and video with a video
The second, third and fourth steps may require half
an hour in total for a video of 8 minutes. Hence, one
can invest most of the time in the first step, namely
the creation of the contents.
In addition to a good balance between quality and
effort, our video creation approach has other advan-
tages: re-editing, correcting errors and translating into
other languages is very fast. One just needs to trans-
late image, script, and re-run the last steps (around 2
hours in total). Perhaps this method could be com-
bined with a pen tablet, a tablet pc or a smartboard of
enough quality to avoid the problems we mentioned
We have tried a new teaching approach using spe-
cially designed videos for self-learning instead of
master classes. The results show that the method is
a success from different perspectives, such as exam
results, and students’ and teacher’s opinions. We be-
lieve that this method may also be useful for other
scientific matters.
Given the success of our videos in the Spanish-
speaking community, we recently translated
them to English to make their contents avail-
able to a wider audience. English version of
videos are posted on youtube and linked from
www.lsi.upc.edu/˜agascon/videostc. We are curious
to see what the impact of this translated material is.
We are also interested in sharing our experience with
other universities. In fact, we have made the videos
publicly available in order to benefit the academic
community in general.
AcademicEarth (2010). http://www.academicearth.org/.
Anderson, R., Dickey, M., and Perkins, H. (2001). Ex-
periences with tutored video instruction for introduc-
tory programming courses. ACM SIGCSE Bulletin,
Bauman, E. and Jurgens, D. (2002). Undergraduate elec-
trical engineering via tutored video instruction. In
Frontiers in Education Conference, 1994. Twenty-
fourth Annual Conference. Proceedings, pages 460–
463. IEEE.
Dale, E. (1969). Audiovisual methods in teaching. Holt,
Rinehart and Winston, Inc., 383 Madison Avenue,
New York, NY 10017.
Day, J., Foley, J., Groeneweg, R., and van der Mast, C.
(2005). Enhancing the classroom learning experience
with web lectures. In Proceeding of the 2005 con-
ference on Towards Sustainable and Scalable Educa-
tional Innovations Informed by the Learning Sciences,
pages 642–645, Amsterdam, The Netherlands, The
Netherlands. IOS Press.
Gibbons, J. F., Kincheloe, W. R., and Down, K. S.
(1977). Tutored Videotape Instruction: A New
Use of Electronics Media in Education. Science,
Handelsman, J., Ebert-May, D., Beichner, R., Bruns, P.,
Chang, A., DeHaan, R., Gentile, J., Lauffer, S.,
Stewart, J., Tilghman, S. M., and Wood, W. B.
(2004). EDUCATION: Scientific Teaching. Science,
KhanAcademy (2010). http://www.khanacademy.org/.
VideoLectures (2010). http://www.videolectures.net/.
CSEDU 2011 - 3rd International Conference on Computer Supported Education