A Digital Application to Assist Basic Education Teachers in the
Interdisciplinary Development of Computational Thinking Skills on the
Math Discipline in Brazilian Learning Context
Erick J. F. Costa, Marcelo G. S. Q. Vitorino, Jo
˜
ao M. L. Medeiros, Cl
´
audio E. C. Campelo
and L
´
ıvia M. R. S. Campos
Systems and Computing Department (DSC), Federal University of Campina Grande (UFCG), Para
´
ıba, Brazil
Keywords:
Collaborative Application, Basic Education, Interdisciplinary Learning Approaches, Computational Thinking,
Math.
Abstract:
Computational Thinking (CT) has skills that can be explored in Math questions and can improve student’s
capability of solving problems. Approaches have been proposed in the literature to support teachers in elab-
orating, sharing, and searching Math questions that stimulated CT skills. However, the wide dissemination
of these approaches has problems to put in practice by teachers. To fill this gap, in this paper, we present a
digital application that aims to help the teachers to work in CT skills through activities of creation, cataloging,
and utilization of Math questions. The validation of the application propose here was conducted with under-
graduate students in Math, who reported that the solution is easy to use and supports and helps teachers put
in practice creating, cataloging, and utilizing Math questions to stimulate CT skills. Also, the application able
the searching and sharing Math questions between teachers easily.
1 INTRODUCTION
Computational Thinking (CT) has become an ap-
proach to stimulating the problem-solving capacity
that has been widely discussed by the community
(Angeli and Giannakos 2020). This approach con-
sists of developing essential Computer Science skills
to design solutions to problems more efficiently and
effectively (Wing, 2006). The skills stimulated by the
CT are described by Barr and Stephenson (2011) as
indispensable, since primary education, for the devel-
opment of problem-solving capacity in students. Two
aspects are commonly considered in several studies
identified in the literature to stimulate these skills:
i) using specific disciplines to stimulate CT skills
(e.g., Programming, Robotics, and Game Develop-
ment) (Rodriguez et al.., 2019 ); ii) second using the
dynamics of the disciplines of the primary teaching
cycle (e.g., Math, Science, and Reading) to stimulate
CT skills by interdisciplinary approaches (Del et al..,
2020 ).
Researchers have evaluated the positive impacts
that approaches conceived from these two perspec-
tives have brought into the classroom in several coun-
tries. These efforts are changing the way traditional
schools work, transforming school life so that stu-
dents can develop necessary skills for any individ-
ual in this new era marked by digital transformations
(Korkmaz, 2018). This scenario has been consoli-
dated in Brazil since the National Common Curricular
Base (Base Nacional Comum Curricular - BNCC).
This document establishes new guidelines for teach-
ing in the country. In teaching Computer Science in
schools, it is possible to highlight, according to the
definitions presented in the BNCC, the general com-
petence “Digital Culture”. This general competence
defines the CT as a piece of essential knowledge for
all students, which should be stimulated interdisci-
plinary, since the initial years at school, together, for
example, with the Math discipline.
When considering the CT’s incentive in conjunc-
tion with Math’s discipline, following the BNCC
guidelines, it is possible to see some efforts that aim
to propose interdisciplinary approaches in this sense
(Palts and Pedaste, 2020). However, computer-based
approaches are considerably more common, such as
those based on computer programming, educational
robotics, and game development. These approaches
have shown significant positive results, but some there
are still significant obstacles to the widespread dis-
Costa, E., Vitorino, M., Medeiros, J., Campelo, C. and Campos, L.
A Digital Application to Assist Basic Education Teachers in the Interdisciplinary Development of Computational Thinking Skills on the Math Discipline in Brazilian Learning Context.
DOI: 10.5220/0010452004750482
In Proceedings of the 13th International Conference on Computer Supported Education (CSEDU 2021) - Volume 1, pages 475-482
ISBN: 978-989-758-502-9
Copyright
c
 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
475

semination of these practices, such as the lack of ad-
equate infrastructure in schools and teachers’ need to
acquire specific CS knowledge.
Our research group conducted previous research,
focused on the interdisciplinary approaches to stimu-
lating CT skills without specific CS knowledge, the
different perspectives of conformity that Math ques-
tions present with the CT skills were evaluated. Based
on these studies, an approach was proposed to en-
abling the CT skills to be incorporated into Math
questions with the aim that the practice of these
new model questions in the classroom improves the
problem-solving ability of primary school students
(Costa et al., 2017). The approach was conceived
from three essential activities: creation, cataloging,
and use of Math questions that stimulate the CT skills;
and was validated from a quasi-experiment, and the
results showed that the questions created, cataloged,
and used, according to the proposal, better stimulate
students regarding the problem-solving capacity.
Although the approach was promising, it is pos-
sible to identify some difficulties around the essen-
tial activities proposed (creation, cataloging, and use).
One of them concerns the impossibility of conduct-
ing cataloging CT skills in Math questions for any
questions. The other is related to the inefficiency in
using the questions produced since the question cre-
ators only have access to what they conceive, making
it difficult for other teachers to share and reuse the
questions created and cataloged.
Because of the difficulties encountered in the stud-
ies conducted previously, the following research ques-
tion was formulated to guide the study presented in
this article: ”How to support and make the process
of creating, cataloging and using Math questions, that
stimulate computational thinking skills, more easily,
efficient and collaborative?”
This work’s objective was to develop a digital ap-
plication that aims to provide a set of resources that
help teachers develop CT skills in combination with
Math questions in response to the research question.
The application was designed to support the activi-
ties of creation, cataloging, and use of questions, pro-
viding strategies for these activities to be carried out
collaboratively - making it easier to share what is pro-
duced and evaluated by teachers within the applica-
tion.
The application proposed in this paper was evalu-
ated by carrying out an extracurricular course with un-
dergraduate students in Math at the Federal University
of Campina Grande (UFCG). At the end of the course,
participants answered a questionnaire about their ex-
perience with the application. The questionnaire’s re-
sponses indicate that the application efficiently assists
the interdisciplinary activities proposed in the litera-
ture (creation, cataloging, and use). Also, the par-
ticipants highlighted high satisfaction with the feed-
back generated by automatic question classifier incor-
porated into the application, using Machine Learning
and Natural Language Processing techniques, for the
questions produced during the course.
The remaining of this paper are organized as fol-
lows: related works are presented in Section 2; the
results in our previous work are presented in Section
3; the research methodology is presented in Section
4; the application design and implementation are pre-
sented in Section 5; in Section 6, we present and dis-
cuss the results of the application validation; and Sec-
tion 7 is dedicated to conclusions and future work.
2 RELATED WORK
The literature highlights some other solutions, based
on digital applications, online tools, and games, to
stimulate the CT skills in the Math discipline and de-
velop the ability to solve problems in an interdisci-
plinary way. In Melo et al. (2018) and Souza et al..
(2018), two gameS models are presented to stimulate
the CT in Math. Both work with logic-based strate-
gies for problem-solving and incorporate CT skills
that involve the abstraction of information and algo-
rithms’ construction.
Barcellos et al. (2016), with the objective of train-
ing teachers on the possibilities of inserting the CT in
the Math discipline, propose a training course using
the Moodle
1
platform, where participants are intro-
duced to activities that use Scratch
2
as a strategy for
incorporating CT into the Math discipline.
The approach proposed by Costa (2014) aims to
train students in Youth and Adult Education so that
they develop the CT through the mobile application
called ForEJA, where strategies are based on exer-
cises. However, these exercises are not focused on
Math.
As it is possible to observe, the literature does not
present, according to the initial exploratory search
carried out, works directly related to the proposal
presented here. However, the works identified show
similarity with the application proposed in this paper,
which introduces CT to stimulate the ability to solve
problems in an interdisciplinary way. However, the
application proposed here differs from the other
solutions due to the range of CT skills addressed and
to the fact that it is applied to Math questions without
1
https://moodle.org/
2
https://scratch.mit.edu/
CSEDU 2021 - 13th International Conference on Computer Supported Education
476

the need for strategies linked to specific CS subjects,
which require specific CS knowledge by teachers, and
adequate infrastructure, by schools, so that they can
be put into practice. Moreover, the application incor-
porates intelligent approaches to creating, cataloging,
and collaboratively using questions.
3 AN APPROACH TO
STIMULATING
COMPUTATIONAL THINKING
IN MATH DISCIPLINE
The literature presents several approaches to stimu-
late CT in the context of education. However, their
wide dissemination is hampered due to the absence
of adequate infrastructure on the part of schools and
teachers’ training to have the specific knowledge in
Computer Science to conduct the proposed activities
in the classroom. In order to overcome these dif-
ficulties, our research group presented an interdis-
ciplinary approach that aims to stimulate CT skills
without the need for specific disciplines of Computer
Science (Costa et al., 2017).
The approach consisted in the adequacy of Math
questions to stimulate a more significant number of
CT skills. In this sense, three activities were pro-
posed concerning the creation, cataloging, and use of
the question: creation activity refers to a step-by-step
that guides teachers on the creation and insertion of
CT skills in Math questions used in the classroom;
cataloging activity consists of guidelines for carrying
out the majority evaluation of the CT skills, based on
manual analyzes, conducted by three experts individ-
ually, for cataloging the CT skills in the questions cre-
ated; and use activity concerns the guidelines regard-
ing the use of the questions conceived with the exe-
cution of the previous activities, in the classroom, by
the teachers.
To conduct the question creation activity, the au-
thors proposed a set of steps to guide how to incorpo-
rate the nine CT skills proposed by Barr and Stephen-
son (2011). From these step by step, it was possible
to produce new questions with greater compliance to
CT skills (questions with more skills associated with
their statement).
To identify the differences present in the ques-
tions produced from the proposed step-by-step and
questions traditionally used in the classroom, the au-
thors conducted a majority evaluation of the ques-
tions’ skills. The process was based on individual
expert evaluations, where, at the end of these evalu-
ations, each skill was cataloged if at least two eval-
uators had identified it. For example, given a ques-
tion, the nine skills were evaluated individually by ex-
perts. If the skill ”data collection” was identified by
two of these specialists, it was cataloged in the ques-
tion. The same procedure was performed for the other
CT skills used in the study. To support the evalua-
tion process, experts used the definitions presented by
Barr and Stephenson (2011), which guide the practi-
cal application of CT skills in Math.
To validate the impact of the proposed approach
on the problem-solving capacity of students was con-
ducted a quasi-experiment (Costa et al., 2017). This
study divided two classes of the eighth year of Brazil-
ian primary education into two groups (experimen-
tal and control). For each group, different question
models were used. The experimental group was used
questions produced from the proposed step-by-step (a
more significant number of CT skills identified in the
primary evaluation and cataloging by specialists), and
for the control group was used legal questions put
into practice by classroom teachers (smaller number
of skills identified in the majority evaluation and cat-
aloging by specialists).
After practices with proposed Math questions over
two weeks, the authors used a questionnaire contain-
ing questions from the PISA (Program for Interna-
tional Student Assessment), an exam that aims to
assess students’ math literacy worldwide and solve
problems final performance analysis tool. The results
showed that Math questions, in greater compliance
with the CT skills, improved the problem-solving ca-
pacity in primary education students. The signifi-
cance (95%) of the results was measured using hy-
pothesis tests. The authors highlighted that the per-
formances obtained by the experimental group were
significantly higher than the performances obtained
by the control group.
The proposed approach was promising, but it is
not viable to manually catalog the CT skills in a more
significant number of Math questions. Also, there was
a mechanism that allowed sharing and reuse, by other
teachers, of the questions that have already been cre-
ated and cataloged. This paper presents a digital ap-
plication design that supports teachers collaboratively
in conducting the essential activities proposed in the
literature to stimulate the CT and math discipline to
support the proposed methodology in literature and
overcome the difficulties mentioned.
4 METHODOLOGY
The methodology followed to conduce this work was
breaking into three steps:
A Digital Application to Assist Basic Education Teachers in the Interdisciplinary Development of Computational Thinking Skills on the
Math Discipline in Brazilian Learning Context
477

• Understanding the approach to stimulating the CT
skills by Math questions to extract the functional-
ities to the concept the application;
• Application implementation based on the func-
tionalities extract on the previous step;
• Validation of the application with Math under-
graduates’ students.
The main functionalities extracted on the first step
are: i) allowing users to create Math questions, which
stimulate CT skills, and add them to an online repos-
itory to facilitate access to these questions in a de-
centralized manner; ii) enabling the cataloging of CT
skills, in each of the questions created in the ap-
plication, to be conducted collaboratively to circum-
vent the difficulties encountered with the manual cat-
aloging of large amounts of questions; iii) making it
possible to use questions from a mechanism for locat-
ing and generating personalized exercise lists.
Based on these functionalities was conducted in
the second step, the application presented here was
conceived from three modules that reflect creating,
cataloging, and using Math questions that stimulate
CT skills. Finally, we conducted the validation pro-
cess, where we set up questionnaires to characterize
the participants and collect their impressions regard-
ing the application proposed here.
5 THE APPLICATION DESIGN
AND IMPLEMENTATION
The application
3
has three main modules that reflect
the functionalities. The creation module allows ques-
tions to be created. At the end of the creation process,
the application provides a feedback preliminary about
the skills present in the elaborated question through
an automatic classifier. The cataloging module, in
turn, allows a newly produced question to be eval-
uated and, in case of acceptance, to be published
for use with the skills cataloged by majority analy-
sis. The search and use module allow, through Infor-
mation Retrieval mechanisms, the questions to be in
the repository and used to compose lists of exercises,
which can be printed from the application.
In the application, two types of actors are consid-
ered: the ordinary user and the administrator. The first
type can take on two leading roles: i) creator, who
takes on the role in creating questions; ii) the eval-
uator responsible for analyzing and identifying the
skills in a newly created question submitted to cat-
aloging. The administrator, in turn, is a user spe-
3
http://compensar-testes.herokuapp.com/
cialization responsible for consulting the three eval-
uations conducted in the majority evaluation process
and, according to the opinion of the evaluating users,
accepting or rejecting the insertion of the questions
created in the collaborative application repository so
that they can be used. Figure 1 presents an overview
of the modules integrated into the application and the
features developed to meet the requirements.
5.1 Creation
In the creation module, the user initially defines
whether the question is objective or subjective and
enters the reference source of the question (e.g., au-
thor, National High School Exam, PISA, among oth-
ers). After defining the initial information, the user
can insert the question’s statement through a text edi-
tor incorporated into the application. This editor al-
lows the question to be formatted according to the
author’s need, inserting images, mathematical expres-
sions (Latex format), among several other text format-
ting options (e.g., bold, italic, and underlined). Af-
ter defining the statement, the user then proceeds to a
new screen where it is possible to select the contents
addressed in the question (e.g., equations, trigonom-
etry, and geometry). Finally, it can insert the answer
alternatives in a template, for objective questions, or
a correction mirror, for subjective questions.
After completing the question elaboration, the ap-
plication automatically presents a preliminary feed-
back of the skills present in the produced question and
initially stores it in the creator’s personal repository.
The feedback is provided from the automatic classifi-
cation by models based on Machine Learning. These
models can analyze the textual content of the ques-
tions’ statements and predict the CT skills present in
the analyzed questions. For training, validation, and
testing of the models, Natural Language Processing
techniques were applied and, as a labeled database,
previously cataloged questions were used, whose CT
skills were identified following the procedures for
majority analysis and cataloging presented in the lit-
erature (Costa et al., 2017). The models incorporated
into the application, according to the results reported
in the literature, had an average accuracy of 92.46 %
for the nine CT skills that were part of the experiments
- which indicates a relevant predictive ability (Costa
et al., 2019).
5.2 Cataloguing
The question cataloging module allows the identifica-
tion of the presence of the CT skills and, according
to the majority evaluation, the acceptance or rejection
CSEDU 2021 - 13th International Conference on Computer Supported Education
478

Figure 1: Overview of the application.
of the questions produced. This model controls the
questions that are inserted in the application. After
creating a question, the author can then submit it to
the evaluation process to publish it definitively in the
application. For that, the question needs evaluations
from its creator and two more evaluating users, to-
taling three evaluations so that, later, it is possible to
apply the majority criterion.
During the evaluation, it is possible to select the
identified skills. In the option ”Tell us more,” the
user can highlight the part of the question’s statement
that refers to the identified skill. Also, it is possi-
ble to inform the confidence level of the evaluation
and general observations about the consistency of the
assessed questions. Specifically, for evaluators other
than the question’s creator, other information is re-
quired, such as suggestions for improvement or minor
modifications - such as spelling errors.
At the end of the individual evaluations, each eval-
uator provides the final publication opinion and indi-
cates whether the question can be published, whether
it needs minor modifications or rejected, and returned
to the creator to apply the necessary corrections. The
final opinion is for the system administrator to be able
to accept or reject the question produced.
The administrator then judges whether to accept
or reject the question produced based on all the iden-
tifications of the skills, proposals for improvement,
and the evaluators’ final opinion. He/she can make
the proposed small changes, accept the publication,
or reject it if the changes are substantial. In case of
rejection, the question is returned to the creator with
the proposed observations to apply them and, if he/she
wishes, resubmit the question for a new round of eval-
uations.
In case of acceptance, the question is registered
in the application’s collaborative repository, and the
skills are mapped according to the majority evalua-
tion. In this process, for each skill evaluated, if at least
two evaluators have identified it, it is cataloged in the
question (it will appear in the searches that the ques-
tion has that skill). The questions accepted for pub-
lication are no longer modifiable by the creator and
serve as input for the automatic classifier, which pro-
vides preliminary feedback, to continue learning and
increasing its predictive capacity, making the auto-
matic opinion more assertive and satisfying for users.
This approach aims to reduce or even eliminate the
need to conduct majority evaluations when the feed-
back more accurate.
5.3 Search and Use
In the search and use module, so that it is possible to
locate the questions accepted and stored in the col-
laborative repository, their statements are stored in
the database using the text indexing property of Mon-
godb
4
(Text Index). This strategy was incorporated
into the application in order to make the search pro-
cess for questions more efficient. Indexing uses ap-
proximation techniques (stemming), which analyzes
the words searched for to reduce them to a root word
and allow the results to include variations of the
rooted word.
The order of return for textual searches is de-
fined using the score, a score assigned to each doc-
ument after a search has been carried out. The index-
ing algorithm classifies the documents best suited to
the search performed in the application with a higher
score. This score defines the order of the questions
that will be displayed on the search screen. Therefore,
documents with the highest score will take prece-
4
https://docs.mongodb.com/
A Digital Application to Assist Basic Education Teachers in the Interdisciplinary Development of Computational Thinking Skills on the
Math Discipline in Brazilian Learning Context
479

dence in sorting. The entire process of (stemming) is
also considered for the skills present in the questions
evaluated.
Search options are presented based on information
related to the content and CT skills to make it possi-
ble to locate the questions stored in the application,
allowing a quick option to return the questions. Also,
advanced searches are allowed, which use the other
information present in the questions (excerpts of text
in the statement, type, source, and author).
The question lists were created to facilitate the use
of the questions produced in the application, thus al-
lowing users to group questions based on their search
criteria. From this functionality, it is possible to cre-
ate lists with different contents, filter the insertion of
skills and make changes according to the need. Also,
the application has a feature that allows the printing
of the question lists and their answer sheets contain-
ing the correct alternatives for the objective questions
or the correction mirror for the subjective ones.
6 VALIDATION OF THE
APPLICATION PROPOSED
The application validation process was carried out
based on an extracurricular course for undergraduate
degree students in Math at the Federal University of
Campina Grande (UFCG) in August 2019. The total
of participants was 24 (9 female and 15 male). The
course was structured in two moments with a total du-
ration of 8 hours: the first moment had the objective
of training the participants so that they could carry
out the activities of creating, cataloging, and using
Math questions that stimulate the CT skills without
the support of the digital application, that is, all activ-
ities were conducted by the participants just using pa-
per and pen; and the second moment, the participants
were introduced to the resources implemented in the
application to support the activities of creating, cata-
loging and collaboratively using the questions. At this
time, the same activities conducted manually were
performed with the help of the application’s features.
During the first stage of the course, held on Au-
gust 29, 2019, some concepts were presented, includ-
ing Computational Thinking (CT), Common Curricu-
lar National Base (BNCC), and Interdisciplinary Ap-
plications of CT in Math. Following that, students
were introduced to the interdisciplinary approach de-
scribed in Section 3 of this paper, which involves the
process of creating, cataloging, and using Math ques-
tions that stimulate CT skills manually, using paper
and pen.
In the second stage, held on August 31, 2019, af-
ter students had performed some activities related to
creating, cataloging, and using Math questions, the
application was introduced. Students were instructed
to carry out the activities during the practice using the
application and available modules. The first module
used was the search module and then the question cre-
ation module. In the end, impressions were collected
about the feasibility of using the application and how
well it supported the activities proposed in the litera-
ture that encompasses stimulating the CT through the
creation, cataloging, and use of Math questions.
At the end of the course, participants answered a
questionnaire on aspects related to the usability of the
application and the feedback generated by the auto-
matic question classifier according to the following
questions: i) ”Is the process of producing Math ques-
tions that stimulate CT skills easy to perform?”; ii)
”The feedback generated by the application about the
CT skills being stimulated in the question of elabo-
rated Math is satisfactory?”; iii) ”Is the process of
searching for questions considering a certain content
in Math and CT skills easy to perform?”.
The results of the questionnaire will be shown be-
low. The participants’ level of satisfaction was high,
and the indication is that the system is straightforward
to use and provides adequate support to the activities
proposed (Figure 2). It is worth highlighting the in-
dication of low ease regarding the creation module by
one of the participants. However, in the open question
asked at the end of the form, the participant indicated
that the application could assist CT in math through
questions and to stimulate the problem-solving ca-
pacity in primary education students to whom the ex-
ercises will be presented. Also, another participant
highlighted the need to improve the question editor,
as it does not allow the direct insertion of tables.
Regarding the automatic feedback for PC skills
generated by the classifier (Figure 3), the level of sat-
isfaction of the participants was high and, when con-
sulting the open responses, it was possible to verify
that not all the CT skills incorporated by them to the
questions were automatically identified. This behav-
ior was already expected since the classifier learns as
more information is provided to them.
The questionnaire indicated that the search tools
were more effective and easy to use the components to
insert information about CT skills and general Math
questions information when compared with the man-
ual process (Figure 4).
In general, it was possible to notice that the stu-
dents showed a great interest in the interdisciplinary
proposal presented. This conclusion was possible
from an open question at the end of the evaluation
CSEDU 2021 - 13th International Conference on Computer Supported Education
480

Figure 2: Distribution of ease of use indications and support
for the teacher of the creation module.
Figure 3: Distribution of satisfaction indications with feed-
back automatic.
Figure 4: Distribution of ease of use indications and support
for the search module teacher.
questionnaire. Some reports of the participants: ”At
first I thought the course would be theoretical and
boring and when we started the course I was quite
surprised. The didactic, the receptivity we were given,
allowed for a very dynamic conversation and course.
In addition, the idAlsoll this is very good and interest-
ing, particularly I had never thought about the ques-
tions in this way, analyzing each skill to be applied.”;
”This training in my view is an ambitious proposal,
which will contribute a lot as a tool to help teach-
ers and create a more specific monitoring for each of
the students.”; ”I found the experience very valid, it
helps to understand better to understand better exer-
cises and problems into the classroom, covering dif-
ferent important skills for teaching.”.
Finally, it was possible to identify that, since the
participants have little knowledge about the inter-
disciplinary practices that consider the CT in Math,
the application allowed their interest in the theme.
These findings could be identified from the open re-
sponses to the questionnaire: ”Yes. I found the appli-
cation to be a handy and interesting platform, even
for students. It may act as a stimulus for students to
study and resolve questions.”; ”Yes. Since the con-
textualization of constructed questions will also allow
for better interaction between different disciplines.”;
“Yes. Being within the scope of work of the skills in-
terconnected with the mathematical, logical thinking,
the application made it much easier to conduct the
activities done manually.”.
Finally, it is worth noting that the cataloging mod-
ule and the list generation functionality were not yet
presented in the application at the course time. There-
fore, it is necessary to validate them in future courses.
Also, the application showed signs that it could sup-
port the activities proposed in the literature that con-
cern creation, cataloging, and use.
The application can provide collaborative and ef-
ficient strategies, considering that the whole process,
in principle, was centered on a single individual re-
sponsible for creating and cataloging the questions
it produced. Another benefit of the application pro-
posed was making the use of CT as a joint approach
with Math accessible to more people, particularly to
teachers interested in applying the methodology in the
classroom.
The preliminary results are encouraging, and it
is possible to implement the proposed improvements
shortly and make the tool available for community
use. Also, by making the application available for
more professionals, it will be possible to gather more
information to assist the application’s validation in all
its dimensions, such as performance, usability, reli-
ability, security, availability, maintenance, and tech-
nologies involved.
7 CONCLUSIONS AND FUTURE
WORK
The CT is considered a strategy to develop the
problem-solving capacity and can be developed in
an interdisciplinary way with the Math discipline, as
we proposed in previous work and strengthened by
the BNCC. To support this proposal, we presented in
this paper an application that provides functionalities
to stimulate the CT’s skills from the creation, cata-
loging, and use of Math questions to help teachers in
the classroom.
A Digital Application to Assist Basic Education Teachers in the Interdisciplinary Development of Computational Thinking Skills on the
Math Discipline in Brazilian Learning Context
481

The results of the application validation on the us-
ability of the question creation and search modules
and the satisfaction generated by automatic feedback
skill showed that it reaches its objective of support-
ing the interdisciplinary practice of stimulating the
CT through Math questions, providing an application
to help teachers to create, cataloging and use. Also,
the application proved to be an option of great im-
portance regarding stimulating the CT in an interdis-
ciplinary way to the discipline of Math, reducing the
need for specific knowledge of approaches linked to
the disciplines of Computer Science by the teachers,
and adequate infrastructure by schools.
As future work is intended to carry out new
courses to validate the application modules regard-
ing the majority evaluation and generation of exer-
cise lists, as they were not yet integrated into the so-
lution when the course described in this paper was
performed. Also, new questionnaires and interview
approaches can be used to identify participants’ real
sentiments to collect more significant results about the
application characteristics.
In our preliminary validation process, we selected
undergraduates students of Math. This selection al-
lows preliminary insights into the application. There-
fore, it is necessary to consider teachers actually in
the classroom in new extracurricular courses for more
significant results about the application efficiency to
help the creating, cataloging, and using Math ques-
tions and CT skills.
With the consolidation of the application, it will
be possible to formalize strategies for using questions.
Their CT skills make it easier for the teacher to iden-
tify the students’ specific difficulties. In this sense,
the teacher will be able to build lists of personalized
exercises to develop the skills with which the students
presented difficulties. Suppose the student had diffi-
culties abstracting important characteristics of prob-
lems. In that case, the teacher could consult the ap-
plication in search of Math questions that specifically
stimulate the ”information abstraction” skill so that
they can be used to develop the student.
REFERENCES
Angeli, C. e Giannakos, M. (2020). Computational thinking
education: Issues and challenges.
Barcelos, T. S. et al. (2016). Formac¸
˜
ao online para o desen-
volvimento do pensamento computacional em profes-
sores de matem
´
atica. In Anais dos Workshops do Con-
gresso Brasileiro de Inform
´
atica na Educac¸
˜
ao, vol-
ume 5, page 1228.
Barr, V. e Stephenson, C. (2011). Bringing computational
thinking to k-12: What is involved and what is the role
of the computer science education community? ACM
Inroads, 2(1):48–54.
Costa, B. V. B. (2014). Pensamento computacional na
educac¸
˜
ao de jovens e adultos: um estudo de caso
utilizando dispositivos m
´
oveis. Monografia, Licen-
ciatura em Ci
ˆ
encia da Computac¸
˜
ao, Universidade de
Bras
´
ılia.
Costa, E. J., Campelo, C. E., and Campos, L. M. S. (2019).
Automatic classification of computational thinking
skills in elementary school math questions. In 2019
IEEE Frontiers in Education Conference (FIE), pages
1–9. IEEE.
Costa, E. J. F., Campos, L. M. R. S., and Guerrero, D.
D. S. (2017). Computational thinking in mathematics
education: A joint approach to encourage problem-
solving ability. In 2017 IEEE Frontiers in Education
Conference (FIE), pages 1–8. IEEE.
del Olmo-Mu
˜
noz, J., C
´
ozar-Guti
´
errez, R., and Gonz
´
alez-
Calero, J. A. (2020). Computational thinking through
unplugged activities in early years of primary educa-
tion. Computers & Education, 150:103832.
Korkmaz,
¨
O. (2018). The effect of scratch-and lego
mindstorms ev3-based programming activities on
academic achievement, problem-solving skills and
logical-mathematical thinking skills of students. MO-
JES: Malaysian Online Journal of Educational Sci-
ences, 4(3):73–88.
Melo, D., de Sousa Pires, F. G., Melo, R., and J
´
unior,
R. J. d. R. S. (2018). Rob
ˆ
o euroi: Game de es-
trat
´
egia matem
´
atica para exercitar o pensamento com-
putacional. In Brazilian Symposium on Computers
in Education (Simp
´
osio Brasileiro de Inform
´
atica na
Educac¸
˜
ao-SBIE), volume 29, page 685.
Palts, T. e Pedaste, M. (2020). A model for developing com-
putational thinking skills. Informatics in Education,
19(1):113–128.
Rodr
´
ıguez-Mart
´
ınez, J. A., Gonz
´
alez-Calero, J. A., and
S
´
aez-L
´
opez, J. M. (2019). Computational thinking
and mathematics using scratch: an experiment with
sixth-grade students. Interactive Learning Environ-
ments, pages 1–12.
Souza, D. et al. (2018). Lightbot logicamente: um game
l
´
udico amparado pelo pensamento computacional e a
matem
´
atica. In Anais do Workshop de Inform
´
atica na
Escola, volume 24, page 61.
Wing, J. M. (2006). Computational thinking. Communica-
tions of the ACM, 49(3):33–35.
CSEDU 2021 - 13th International Conference on Computer Supported Education
482