Teaching of Programming - An Educational Performance Booster for
Students at Economically Disadvantaged Schools
Paulo Brito
1
, J. Antão B. Moura
2
, Joaquim Honório
2
, Marcelo Barros
2
and Igor Vieira
2
1
Graduate Program in Computer Science, Federal University of Campina Grande (UFCG), Brazil
2
Systems and Computing Department, Federal University of Campina Grande (UFCG), Brazil
mbarros@computacao.ufcg.edu.br
Keywords: Teaching of Programming, Educational Performance, Computer Science Education, Computational
Thinking.
Abstract: The technological advances made in recent years bring with them the importance of introducing computer
courses in the school context, and with this, a need for digital inclusion of students from economically
disadvantaged schools. This paper reports on research to introduce programming in such school curricula
and to evaluate possible benefits of such introduction for students’ motivation towards learning in general.
The research was based on the creation of a methodology for designing and offering programming courses
at public and private schools and to verify possible correlations between students’ performance and schools’
economic scenarios they find themselves in. Preliminary results indicate that students who participated in
the courses at schools with economic restrictions and inferior quality of IT infrastructure may still present
better results in the courses. In addition, there is evidence of gains in their motivation towards learnig other
subjects. The paper details results, analyzes causes for them and illustrates and explores implications for
participating students at schools operating across the economic spectrum.
1 INTRODUCTION
With globalization and the technological advances
that have occurred in recent years, it is becoming
increasingly important to incorporate subjects in the
curriculum that meet the technological needs of the
present century (Pagani, 2015). Thus, the teaching of
programming at the levels of basic education has
grown worldwide due to its benefits to the learning
process. The benefits include general pedagogical
aspects, such as the development of logical
reasoning, new ways of thinking about a problem
and interdisciplinary learning, among others
(Barbosa, 2017).
However, there are many difficulties to the
teaching of programming in the educational
structure in developing countries. Some of theses
difficulties, like: insufficiente number of computers,
lack of knowledge/skills of the teachers, not enough
copies of sotware and difficulty to integrate in
instruction not only also affect some developed
countries as also the teaching of other subjects,
creating obstacles for the implementation of a good
infrastructure of Information and Communication
Technology (ICT) (Pelgrum, 2001). Furthermore,
these difficulties are aggravated, especially when it
comes to education in less favored or “economically
disadvantaged” schools (Duarte, 2007). Here an
“economically disadvantaged school” is a school
that is poor, i.e., it faces problems and difficulties
due to lack of (enough) money and other economic
resources - such as infrastructure in general, but IT
in particular - and it is usually located in a low-
income community. That is the case of public
elementary, middle or high schools run by the
government in Brazil. The economic disadvantage
typically exists between a private and a public
school; but it may also appear amongst public
schools: one school may be more economically
disadvantaged than another.
Economically disadvantaged schools are often
the result of various factors and/or the interactions of
these factors. In this work, the following indicators,
when evaluated in want, are assumed to characterize
a disadvantaged school: (i) number of functioning
computers available to students; (ii) existence of
computer classes; (iii) the quality of technological
resources such as (old, little memory) computers and
Brito, P., Moura, J., Honório, J., Barros, M. and Vieira, I.
Teaching of Programming - An Educational Performance Booster for Students at Economically Disadvantaged Schools.
DOI: 10.5220/0006785101550164
In Proceedings of the 10th International Conference on Computer Supported Education (CSEDU 2018), pages 155-164
ISBN: 978-989-758-291-2
Copyright
c
2019 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
155
Internet access; (iv) quality of the school’s overall
infrastructure.
Previous research concentrated on investigating
programming teaching strategies and indicated
promising results (Barbosa, 2017) (Pina and Rubio,
2017). Grover, Pea and Cooper (2016) studied
factors (e.g. prior technology experiences and
student demographic information) that may affect
results of programming courses. While studies like
Papadakis, Kalogiannakis, Orfanakis and Zaranis
(2017) pointed out some of the mainly barriers
encounter by novices in these courses. However, the
literature is short of works that assess the impact of
economical resources of schools on the performance
metrics of programming courses – be theses metrics
specific to programing or generic in academic terms.
Such an assessment is the object of this paper.
This paper discusses research results on the
impact the teaching of programming as a curricular
subject may have on the level of learning the subject
but also as a performance driver for the overall
academic motivation of students at economically
disadvantaged schools. For that, two research
questions (RQs) are of interest: one that relates
students’ performce on the introductory
programming course to their school facilities; and,
the other that explores benefits to students’ academic
behavior in general. To address these questions,
albeit preliminarly, experiments have been designed
based on the code.org platform (Kalelioglu, 2015)
for the teaching of programming at private and
public schools with varying degrees of economic
disadvantages in Northeastern Brazil. This paper
reports on the methodology for the design,
application and evaluation of results of a course on
introductory programming concepts.
Besides contributing preliminary evidence to
answer the RQs, this paper may open new research
opportunities on the topic of teaching of
programming and its implications.
The remainder of this paper is structured as
follows. Section 2 highlights basic concepts and
reviews related work. In section 3, the methodology
and the models used in the research are presented.
Section 4 offers and analyses results of the
considered introduction to programming course.
Finally, section 5 presents’ conclusions and explores
future work opportunities on the topic.
2 RELATED WORK
Currently, most students from any school spend a
large part of their free time playing computer games.
Hence, integrating games into school subjects can
increase student interest by providing opportunities
for them to learn while having fun
(Grivokostopoulou, Perikos and Hatzilygeroudis,
2016). Computer games can be used to teach almost
all areas of computer science, and researchers point
out that they could effectively provide the most
interesting learning environments for knowledge
acquisition and construction (Sung and Hwang,
2013).
Some works have investigated how to apply
game concepts to programming teaching. Pina and
Rubio (2017) have shown that the use of educational
systems based on games brings a significant
advantage to students, improving motivation and the
general interest in learning. Ibáñez et al. (2014) also
investigated the use of a strategy based on games for
the teaching of programming basic concepts to
undergraduate students. Barbosa (2017) states that,
in programming logic classes for 10-year-olds,
improvements were made in logical reasoning,
where students understood algorithm concepts and
variables by analogies with situations that they have
already experienced, such as following a cake
recipe. However, the teaching-learning process of
programming is not a trivial task. Rodrigues (2002)
highlights several problems that may affect the
teaching and learning process of algorithms and
programming, including: i) difficulty in making
students develop logical reasoning when they are
accustomed to memorize content; and, ii) lack of
motivation of students generated by their lack of
preparation and discouragement when they believe
that the course presents difficult obstacles to be
overcome. Grover, Pea, and Cooper (2016) studied
factors related to students, such as past experiences
and demographic data that influence learning
outcomes.
However, although the above related papers
discuss strategies, benefits and success factors of
programming teaching, little has been discussed
about the possible relationship between students’
academic performance in general (and on
introductory programming concepts in particular)
and conditions and resources made available by their
schools for a programming course.
This work extends previous research by
analyzing the influence an introductory
programming course may have on the overall
academic performance of students at economically
challenged schools.
CSEDU 2018 - 10th International Conference on Computer Supported Education
156
Table 1: Characterization of participating schools.
School Functioning
Computers
Total
Number
of
Students
in
Proposed
Course
Proportion of
Computers for
Each Student
Regular
Computer
Classes (1
– Yes; 0 –
No)
Quality of
Technology
Resources
(Computer
and Internet
Access: 1-
Very Bad to
5- Very
Good)
Quality of
School
Infrastructure
in General
(1-Very Bad
to 5- Very
Good))
Overall
Indicator
(Ranking
Index -
RI)
Private
School
40 11 3.63 1 5 5 10.04
School A 10 12 0.83 1 4 4 6.12
School B 8 16 0.50 0 5 4 6.12
School C 9 17 0.52 1 4 3 5.39
School D 14 18 0.77 1 3 4 5.32
School E 14 20 0.70 0 3 3 4.27
School F 14 21 0.66 0 3 3 4.24
3 METHODOLOGY
The objective of the research reported here is to
analyze the impact of teaching a programming
course as a code.org application. More specifically,
the research aims to investigate how students from
econmically disadvantaged participating schools
behave when they come in contact with this new
pedagogic resource.
3.1 School Characterization
Participating schools were ordered from the least to
the most disadvantaged (please, see Table 1). A total
of 7 schools were included in the study: one private
school and 6 public schools (A to F).
The characterization was done using the criteria
of Table 1, each of which received a score according
to the associated degree of importance (I): 1 (very
low importance), 2 (low importance), 3 (medium
importance), 4 (high importance) and 5 (very high
importance). A brief description of each criterion
follows.
Number of functioning computers
available to students: the number of
computers that are in perfect working order
and are intended for student use. Although
websites such as QEdu.org provide data on
the number of computers in public schools
in Brazil, the actual number of computers
in operation does not match that informed.
The amount of computers that were seen as
available during the course was used
instead. This characteristic is defined as
being very important because the number
of computers can directly impact the
performance of programming classes.
Computer classes: Earlier work, such as
Grover's, Pea and Cooper (2016), pointed
out that previous experiences students may
have had with technology may have a direct
effect on learning. Thus, as a result of what
was seen during the course, this
characteristic was defined as being of
medium importance. Because it is a
categorical variable, we assign scores (N)
considering the following possibilities: 0 -
the school does not provide computer
classes and 1 the school provides computer
classes.
Quality of technological resources: The
quality of resources refers to the condition
of the internet access and computers. As
with the number of computers, this feature
may be a result of a lack of school
resources and may have a direct impact on
teaching programming. Therefore, this
characteristic is assigned a great
importance. The score for this criterium
corresponds to the quality of technological
resources: 1 (very bad quality), 2 (bad
quality), 3 (medium quality), 4 (good
quality) and 5 (very good quality).
Teaching of Programming - An Educational Performance Booster for Students at Economically Disadvantaged Schools
157
Quality of School Infrastructure: The
quality of school infrastructure refers to the
perceived condition of all school assets,
from tables and seats, to the structural state
of classrooms and computer lab. This
characteristic may be considered of
medium importance. Its score corresponds
to the quality of the school infrastructure: 1
(very low quality), 2 (low quality), 3
(medium quality), 4 (good quality) and 5
(very good quality).
RI: the overall evaluation of the previous
criteria – which corresponds to the ranking
index of a school, obtained according to
quation 1. In Table 2 a list of criteria
weighted by the degree of importance will
serve as a basis to define how much the
school is disadvantaged.
Equation 1: Summary of criterion assessment (RI)
=
∑

where
:
:
:ℎ
(1)
Table 2: Weight of each criterion used.
Description of
criterion
I: Importance of
criterion or
Weight
Proportion of
Computers for
Each Student
3
Computer classes 2
Quality of
Technology
Resources
3
Quality of
Infrastructure
School
2
3.2 Research Questions
Two research questions (RQs) are of interest here:
RQ
1
: Are there any performance differences
between the learning levels of programming
among students at schools of varying
economic capabilities?
RQ
2
: Does the teaching of programming
contribute to students' motivation towards
other subjects (and not only mathematics or
logic)?
The following hypotheses are associated to RQ
1
:
Null hypothesis, 1H
0
: there is no difference
between levels of programming (measured by the
number of completed programming levels) among
students from schools of different economic
capabilities. The null hypothesis may be formalized
as follows:

:μ

=μ

Where the notation
μ

represents the mean of
the performance results (scores) by participating
sudents from school in exams or tests after
concluding the Introduction to Programming Course.
Using similar notation, one may define the other
hypotheses as follows.
Alternative hypothesis, 1H
1
: there is a significant
difference between levels of programming learning
(measured by the number of completed levels)
among students from schools of different economic
capabilities. The alternative hypothesis is formalized
as follows:

:μ

≠μ

And for RQ
2
:
Null hypothesis, 2H
0
: teaching of programming
does not contribute to student motivation towards
other disciplines in general (not just math and logic).
The null hypothesis is formalized as follows:

:μ

=μ

Alternative hypothesis, 2H
1
: teaching of
programming contributes to student motivation
towards learning other disciplines (not just math and
logic). This alternative hypothesis is formalized as
follows:

:μ

≠μ

For the development of the work, the underlying
research is of mixed nature, i.e., the quantitative
results from the case study with code.org in terms of
score on tests on programming skills and knowledge
that were acquired with the taught course are
complemented with interviews with teachers from
participating schools. These interviews served to
CSEDU 2018 - 10th International Conference on Computer Supported Education
158
estimate students’ post-course academic motivation
and were conducted through telephone calls, based
on open-ended questions (please refer to subsection
3.5).
3.3 Code.org Course Program
Code.org is a non-profit organization that provides a
programming teaching platform with the goal of
promoting and teaching programming to people of
all ages. The site includes free coding classes and
the initiative also targets schools in an attempt to
encourage them to include more computer classes in
their curricula. The main objective of a code.org
course is to find solutions to problems that are
presented graphically and the student are supposed
to provide solutions in the form of a set of
sequenced blocks, for example.
The founders of code.org believe that Computer
Science should be a discipline alongside
conventional disciplines such as mathematics,
physics, and biology (Pantaleão, Amaral and Silva,
2017). For each new activity on Code.org there is
always a video at its beginning, with the purpose of
motivating and explaining how to carry out the
activity. For children who do not know how to read,
activities are presented as images and symbols
which are linked to the daily life of the child.
An illustration of a course activity is given in
Figure 1. In this activity, the student must assemble
a sequence of blocks corresponding to the steps
necessary for Angry Bird to reach the green pig.
Figure 1: Code.org platform screen that shows the
organization of a task to be solved by the student.
The methodology for applying the introduction
to programming course consisted of five steps (as
shown in Figure 2). The first one evolved around the
planning of the basic concepts to be worked out by
the students. The concepts were chosen according to
the level of difficulty for learning, considering the
age of the students and the respective school year
they were at. The importance of learning these basic
programming concepts for advanced programming
was also considered. Five concepts were chosen in
step one:
Algorithms: An algorithm is a sequence of
“actions” that must be followed by the
computer to solve a certain task. This sequence
of actions, specified by the human programmer,
much like a cake recipe, will produce a solution
for the problem at hand.
Sequencing: a set of commands of an
algorithm, which are organized in a sequence,
in an attempt to solve a certain problem or task.
On the Code.org platform, the user orders a
sequence of instructions / actions by dragging
blocks of commands and arranging them one
below the other, thus forming the algorithm.
Repetition Block: a set of blocks repeated in an
algorithm. When the student reaches this
learning stage, the alternatives will be studied to
improve the produced algorithm, introducing
blocks that replace the repeated blocks,
reducing the size of the algorithm, but having
the same purpose and results when executed.
Debugging Algorithms: the process of finding
and fixing defects of a particular algorithm or
software. Typically, these defects produce an
unexpected (wrong) result or do not execute
when compiled. In the proposed course, some
problems are presented to the student already
with pre-defined solutions, but with errors. The
student must debug the algorithm to find and fix
the defects.
Condition Block: a flow deviation control
structure present in programming languages that
performs different computations or actions
depending on whether the condition is true or
false. In the planned course, the "if-then (-
soon)" structure was worked through blocks
that represent this structure.
In the second stage, a mapping of the schools
that could participate in the project in the city of
Campina Grande-Paraíba, Brazil was carried out.
Next, the mapped schools were visited and evaluated
to check whether they had the necessary
infrastructure (e.g. computers and internet access) to
support the introduction to programming course.
Given a school’s infrastructure was considered
adequate, the code.org based course was then
offered for students at that school.
Teaching of Programming - An Educational Performance Booster for Students at Economically Disadvantaged Schools
159
Figure 2: Methodological process for the courses.
3.4 Evaluation Procedure
After completing the course, students underwent a
quantitative assessment using the data stored on the
Code.org platform. In addition, a qualitative post-
course questionnaire about their (new) academic
behavior was answered by teachers of the
participating schools. Indicators such as motivation
and concentration were used as proxies for
“academic behavior”.
3.4.1 Quantitative Evaluation
When solving problems in the course based on the
Code.org platform, two main metrics were collected:
Number of Completed Levels and Lines of Code.
The Number of Completed Levels corresponds to
the number of problems solved during the course.
Lines of Code corresponds to the amount of code
written when solving problems in the course.
Consequently and typically, the higher the number
of completed levels, the greater the number of lines
of code written. However, it should be noted that the
number of lines of code may vary due to the solution
used to solve a problem.
3.4.2 Post-course Qualitative Evaluation
After the course was taught at a school, the open
questions in the questionnaire of Table 3 were used
with the objective of verifying possible
improvement in students’ academic behavior. Such
behavior was summarized by metrics that include:
class attendance, study motivation, and participation
in classes.
Table 3: Questionnaire for post-course evaluation.
Id Questions
Q1
Do participating students' appear more
motivated towards academic activities in
general?
Q2
Has the aggressiveness (bullying) by
students who took the course reduced?
Q3
Have students become more motivated to
attend computer classes?
Q4
Has the concentration of students increased
during class?
Q5
In general, did the class improve compared
to classes that did not participate in the
course?
Q6
Do students participate more in the classes
after the course?
4 RESULTS AND ANALYSES
Courses on Intrduction to Programming based on the
code.org plataform were carried out in six public
schools (A to F) in the city of Campina Grande, and
at one private school in the city of João Pessoa. Both
citieis are in the state of Paraíba in Northeastern
Brazil. On average, each class had 12 students who
participated in the course, with a total workload of 3
hours. Participating students were typically 11 years
old and were enrolled in the 5
th
year of elementary
education. Each course lasted 3 hours and each
lesson lasted for an hour and a half. The courses
were taught by one of the authors of this paper. In
some schools, existing IT teachers participated as
assistants to the course's instructor and, at the same
time, learned how to use the platform to continue
this project in their schools. The research questions
are addressed next.
Research Question 1: are there any performance
differences between the learning levels of
programming among students at schools of varying
economic capabilities? Figure 3 and 4 offer some
insght which might allow one to answer this RQ1
preliminarly.
CSEDU 2018 - 10th International Conference on Computer Supported Education
160
Figure 3: Boxplot of completed levels over all classes.
From Figure 3, one may observe that:
a) Classes from all participating schools, even
from those with economic restrictions,
finished the course with the Complete or
Superior Levels completed. Despite having
limited infrastructure and lower quality
internet access, public school students
achieved good results.
b) Two outliers with high values of completed
programming levels (to the right in the
figure) of classes from schools E and F
signal students who were able to stand out
and, during the short period of time the
course lasted for, obtained higher levels of
programming proficiency. On the other
hand, the outlier with below-average value
from school E is due to students who were
later identified as having trouble reading or
concentrating. Reading difficulties, being a
direct (but not sole) responsability of
educators, should be addressed as early as
possible or the affected individuals may risk
having their negative effects spread to other
disciplines.
c) The most economically favored school of
the sample – the Private School – does not
show better results than its public,
economically disadvantaged counterparts.
Human resources - i.e., students in this case
– may compensate for deficiencies in other
resources, up to a point (one cannot do real
programming without machines). Figure 4
may shed more light into this observation.
Figure 4: Averages of programming levels completed by
classes (bars denote 95% confidence interval).
Figure 4 highlights the means of levels
completed in each school, and their respective 95%
confidence intervals. As can be observed, schools A,
C, F, E and Private School present overlapping
confidence intervals. Thus, it is not possible to
isolate significant statistical differences among
them. However, one can state with 95% certainty
that School B has a significantly larger number of
completed levels than the other schools, except for
school A. Similarly, school E shows a significant
difference from schools C and A, having a lower
average compared to them.
It is important to note that the students of school
B - even though they have fewer computers
available and their school provides no regular
computer classes - come out ahead in terms of better
metrics than those of their peers at the other schools
(except for school A whose 95% confidence
interval’s higher half overlaps with that for school
B). This leads one to especulate that students, having
come into contact with the novelty of learning
programming, may have worked harder to explore
the “unusual” opportunity. The quality of
infrastructure and of the technological resources
may have also contributed for the number of levels
completed by B students. Note from Table 1 that
school B, even though it has fewer computers per
capita, presents better structural and equipment
quality. This suggests that it pays to invest in quality
– even in the case of a modest investment. Further,
the 0.5 computer per student ration naturally led to
“pair programming” at school B (i.e., two students
per machine doing programming assignments). This
agile programming (Abrahamsson et. al., 2017)
recommended practice does indeed promote
academic benefits in terms of discussing and
adopting better solution alternatives.
Teaching of Programming - An Educational Performance Booster for Students at Economically Disadvantaged Schools
161
One can also observe that according to Figure 4,
there is no significant relationship between the more
favored (or less disadvantaged) schools and the
number of completed programming levels. With
95% confidence, the school with better resources
(i.e., the private school) presents an average similar
to that of its more disadvantaged counterparts.
Figure 4 suggests that students from too much
economically disadvantaged or favored schools will
do worse (operating in extremes is usually not a
good strategy). Indeed, students from schools E and
F and the private school show programming
proficiency results that could be 20 to 40% less than
their peers from other schools. It pays to invest
(even a little) in quality for that may motivate
students to excel, knowing they will be able to
accomplish what they have set out to do. Motivation
is further explored in RQ2.
Research Question 2: Does the teaching of
programming contribute to students' motivation
towards other subjects (and not only mathematics or
logic)?
To answer RQ2, the post-course questionnaire of
Table 3 was presented to eight teachers of the seven
participating schools and yielded the following
results. It is worth noting that a class at the studied
schools usually has just one teacher, or at most two,
who teaches all the curricular components of the
school series. All interviewed teachers (six females;
two males) had at the time, more than 5 years of
experience teaching children and young people.
Interviewed teachers stated that students who
took the introduction to programming course are
more likely to participate in classes. Teachers also
perceived a reduction in bullying (question 2 in
Table 3) as the consolidation of answers in Figure 5
illustrates. Bullying is typically, a school
problem (Oliveira et al., 2015). One may speculate
that its causes include low (beneficial) occupation of
pootential bullies. Interviewed teachers agree that
students who took the programming course seem to
become more focused and intrested on their studies.
That may have led to the perception of reduced
bullying at school. But that is not explicitly endorsed
by (Oliveira et al., 2015) - although it may be
covered in the indicated "other causes" - nor can one
claim here to have produced evidence for such
cause-effect relationship. For that, furhter and longer
lasting studies are needed.
Figure 5: Qualitative research Q2 results.
Teachers also noticed that students have become
more motivated to learn about computer science,
regardless of the subject taught in the lab. Interest in
programming has increased and students ask for
classes on the Code.org Platform. Interviewed
teachers said they noticed a greater concentration on
academic activites by participating students
particularly in regards to IT-related subjects (at
schools where they are taught).
Regarding overall academic performance of
participating students, the teachers commented
improvements were across all subjects but more
noticeable in mathematics. Teachers believe having
introduction of programming as regular curricular
discipline with a weekly workload in the computer
lab will bring benefits to all involved.
The answers to the questionnaire and results in
Figures 3 and 4 provide (early) evidence the
designed and taught course will positively impact
participating students’ academic performance. For
that there should be some investment in the quality
of the school’s supporting sinfrastructure and IT
resources. Such an investment seems to cause better
results overall, across the curriculum, even at
economically disadvantaged schools. However, it is
important to notice that, at the stage of the research
reported in this paper, the post-evaluation discussed
here was solely carried out with teachers. A more
thorough and comprehensive longitudinal (i.e.,
across several subjects in the school curriculum)
evaluation of participating students' gains with the
programming course needs to be done in the future
for sounder and further claims.
87%
13%
Q2: Has the aggressiveness (bullying) by the
students in the course reduced?
YES NO
CSEDU 2018 - 10th International Conference on Computer Supported Education
162
5 CONCLUSIONS AND FUTURE
WORK
This article discussed the entire process for the
creation and execution of an introduction course on
programming applied in private and public schools,
with the objective of answering two research
questions: i) Are there any performance differences
between the learning levels of programming among
students at schools of varying economic
capabilities? And, ii) Does the teaching of
programming contribute to students' motivation
towards other subjects (and not only mathematics or
logic)?
Through the use of the code.org platform,
metrics were collected that allowed exploration of
correlations between student performance and issues
related to school economic constraints. A qualitative
post-course research was also carried out, with the
objective of analyzing gains of motivation and
performance of students in the classroom.
The results, both with the analysis of the data of
the platform and with the post-course research
carried out with the teachers, allowed one to
consider some interesting aspects. In relation to the
data obtained from the platform and with the
methodology applied to classes, the students of
public schools, even in an environment with budget
constraints and inferior infrastructure, compared
rather well against their peers from the private
school with better economic conditions. This
suggests that the platform environment allows for
effective study, making students who never had
contact with programming, solve problems
presented during the course. In addition, the
curiosity of the students of the public schools for
novelty, that is, for being immersed in a course that
they never had the opportunity before, favored their
engagement and their good results at the end of the
course.
The post-course research conducted with the
teachers indicated that participating students became
more motivated towards their studies and more
participative and more concentrated in classes.
These findings and possibly new ones will have to
be verified further with new longer lasting validation
studies, with the introduction of a course or even a
programming discipline with a longer workload. The
research also suggests that if larger investments are
made in public schools, with the introduction of a
programming discipline or for improvement of
available laboratory infrastructure, their students
may achieve better performance.
The project brought contributions to the school
environment. Considering the results of public
school classes, one may state that programming
courses can be an important tool for digital inclusion
in schools that have financial restrictions. In the
scope of the research, contribution came about by
showing that, applying a pedagogic methodology
capable of inserting the student in an environment
that arouses her or his curiosity and motivation,
good results will result.
As future work, one can anticipate additional
experiments to collect more meaningful statistics; to
select topics, concepts and programming facilities to
be included in course material; and to investigate
their impact on the motivation of disadvantaged
students towards education. Additional experiments
may start with pilot projects to introduce the study of
programming as a discipline in the curriculum.
REFERENCES
Abrahamsson, Pekka; Salo, Outi; Ronkainen, Jussi;
Warsta, Juhani; Agile Software Development
Methods: Review and Analysis. 2017. Technical
Research Centre of Finland.
Barbosa, Fernanda de Arruda Campos. 2017. Computer
Lessons in Early Childhood Education and Elementary
Education: Importance and Benefits for Integral
Formation. Technological Interface Magazine, pages
9-20.
Duarte, Clarice Seixas. 2007. Education as a Fundamental
Law of Social Nature. Educ. Soc., pages 691-713.
Grivokostopoulou, Foteini; Perikos, Isodoros;
Hatzilygeroudis, Ioannis. 2016. An Educational Game
for Teaching Search Algorithms. In Proceedings of the
8th International Conference on Computer Supported
Education, pages 129-136.
Grover, Shuchi, Roy Pea, and Stephen Cooper. 2016.
Factors influencing computer science learning in
middle school. Proceedings of the 47th ACM technical
symposium on computing science education. ACM.
Ibáñez, María-Blanca; Di-Serio, Ángela; Delgado-Kloss,
Carlos. 2014. Gamification for Engaging Computer
Science Students in Learning Activities: A Case
Study. IEEE Transactions on Learning Technologies,
pages 291-301.
Kalelioglu, Filiz. 2015. A new way of teaching
programming skills to K-12 students: Code.org.
Baskent University. Computers in Human Behavior,
pages 200-210.
Oliveira, Wanderlei Abadio de et al. 2015. The causes of
bullying: results from the National Survey of School
Health (PeNSE). Revista latino-americana de
enfermagem, pages 275-282.
Pagani, Mario Mecenas. 2015. The Insertion of
Technologies in Luddical and Recreational Activities
Teaching of Programming - An Educational Performance Booster for Students at Economically Disadvantaged Schools
163
in the 4
TH
and 5
TH
Years of Fundamental Teaching.
FAEMA Scientific Journal, pages 90-106.
Pantaleão, Eliana; Amaral, Laurence Rodrigues; Silva,
Glaucia Braga. 2015. An approach based on Robocode
environment for teaching programming in high school.
Brazilian Journal of Computers in Education, Vol 25.
Papadakis, S.; Kalogiannakis, M.; Orfanakis V.; and
Zaranis, N. 2017. The appropriateness of scratch and
app inventor as educational environments for teaching
introductory programming in primary and secondary
education. International Journal of Web-Based
Learning and Teaching Technologies (IJWLTT),
12(4), 58-77.
Pelgrum, W. J. 2001. Obstacles to the integration of ICT
in education: results from a worldwide educational
assessment. Computers & education, 37(2), 163-178.
Pina, Alfredo; Rubio, Gabriel. 2017. Using Educational
Robotics with Primary Level Students (6-12 Years
Old) in Different Scholar Scenarios: Learned Lessons.
In Proceedings of the 9th International Conference on
Computer Supported Education, pages 196-208.
Rapkiewicz, Clevi Elena; Falkembach, Gilse; Seixas,
Louise; Rosa, Núbia dos Santos; Cunha, Vanildes
Vieira; Klemann, Miriam. 2006. Pedagogical
Strategies in The Teaching of Algorithms and
Programming Associated with The Use of Educational
Games. New Technologies in Education Magazine,
pages 1-11.
Rodrigues, M. C. 2002. How to Teach Programming?
Informatics – Newsletter, Year I n° 01, ULBRA.
Silva, Teresa Lúcia. 2009. The financing resources
decentralization as a democractic management
inductor. Study about the public schools in São Carlos.
Digital Library of Theses and Dissertations of USP.
Sung, HY; & Hwang, GJ. 2013. A collaborative game -
based learning approach to improving students
learning performance in science courses. Computers &
Education, pages 43-51.
CSEDU 2018 - 10th International Conference on Computer Supported Education
164