Manna Methodology: A Novel Approach to Education 5.0 Through

Learning and Socio-Emotional Assessment in IoD Bootcamps

T. T. Madrigar, R. Calvo and L. B. Ruiz Aylon

Manna Ecosystem, State University of Maringá, Brazil

Keywords:

Education 5.0, Exponential Technologies, Socio-Emotional Skills, Internet of Drones (IoD), Imersive

Bootcamps, Personalized Pedagogy, Machine Learning, Learning Styles, Artiﬁcial Intelligence, 21st Century

Skills.

Abstract:

This study explores the practical application of Education 5.0 through a methodology that integrates expo-

nential technologies, emphasizing the Internet of Drones (IoD) to enhance students’ technical and socio-

emotional competencies. Conducted within the Manna Ecosystem—Brazil’s largest platform for teaching,

research, and innovation in Exponential Technologies—the study evaluates the impact of personalized peda-

gogical approaches in public schools. The proposed methodology utilizes David Kolb’s learning style model

(Accommodator, Converger, Assimilator, and Diverger) and speciﬁc socio-emotional competencies, including

emotional awareness, emotional regulation, and relationship skills. Based on these learning proﬁles and par-

ticipants’ socio-emotional competencies, teams were organized with complementary proﬁles, using Artiﬁcial

Intelligence to create personalized groups that promoted greater engagement and synergy, enhancing perfor-

mance in innovation activities. Four case studies were conducted, comparing experimental groups formed

with the personalized methodology and control groups organized randomly, through diagnostic, formative,

and summative assessments. Statistical analyses revealed that the experimental group showed signiﬁcantly

superior performance compared to the control group, highlighting the potential of the methodology in creating

an interactive and collaborative learning environment. These ﬁndings reinforce the relevance of integrating

exponential technologies, such as IoD, with active and adaptive teaching methodologies, contributing to the

advancement of Education 5.0 and preparing students for the challenges of the 21st century.

1 INTRODUCTION

The digital transformation is reshaping how we live,

work, and learn, blurring the boundaries between

online and ofﬂine into a paradigm Floridi describes

as "onlife." In this new reality, technology seam-

lessly integrates into daily life, creating profound

changes (Floridi, 2014). This scenario expands ed-

ucation’s role in cultivating skills that transcend tech-

nical knowledge, emphasizing socio-emotional com-

petencies essential for today’s world. In response, Ed-

ucation 5.0 emerges as an approach that blends tech-

nological advancements with a humanistic focus, fos-

tering holistic development and preparing individu-

als to tackle the 21st century’s complex challenges.

(Hussin, 2018). This educational perspective extends

beyond keeping pace with technological innovation.

It leverages exponential technologies, such as Arti-

ﬁcial Intelligence (AI), the Internet of Things (IoT),

and the Internet of Drones (IoD), to enrich learning

experiences. These technologies foster personaliza-

tion, protagonism, and collaboration in inclusive, in-

terconnected educational environments, aligning with

the onlife experience and addressing contemporary

societal needs. (World Economic Forum, 2020).

Recent literature emphasizes Education 5.0 as

an approach that combines technological advance-

ments with human development, promoting compre-

hensive training for a world in transformation (Xu

et al., 2018), (Popenici and Kerr, 2017), (Torres

et al., 2019). This approach aims to go beyond

technical skills, incorporating socio-emotional com-

petencies, creativity, and the capacity for innova-

tion, thus preparing citizens to proactively and col-

laboratively tackle contemporary challenges (Salmon,

2019), (Hussin, 2018), (Ford, 2021).

For the Manna_Team (Manna team of scientists),

one of the pioneering groups in this area, Education

5.0 represents the synthesis of Exponential Technolo-

gies and Exponential People, characterized by happi-

504

Madrigar, T. T., Calvo, R. and Aylon, L. B. R.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps.

DOI: 10.5220/0013220900003932

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 2, pages 504-515

ISBN: 978-989-758-746-7; ISSN: 2184-5026

ness, creativity, social intelligence, and the capacity

to drive innovation, contributing to collective well-

being. The union between individuals and exponen-

tial technologies enables the development of expo-

nential schools and universities, forming a generation

of "5.0 Citizens," equipped with hard skills and soft

skills, ready to drive innovations with a positive im-

pact on society. In this sense, the culture of innova-

tion is nurtured from preschool education and extends

to Elementary School, High School, Technical Educa-

tion, and Higher Education, fostering lifelong innova-

tion skills and adaptation, which continuously foster

new businesses and solutions for environmental, so-

cial, and governance challenges.

However, Education 5.0 faces signiﬁcant chal-

lenges, particularly in bridging the gap between

universities, notably in engineering and comput-

ing courses—responsible for hard skills develop-

ment—and public schools, which are run with dedica-

tion by pedagogues and educators. The Manna_Team

addresses this challenge directly by integrating prac-

tical kits of Artiﬁcial Intelligence, IoD, IoT, Robotics,

and Games into public schools, promoting soft

skills development through an innovative method-

ology known as MannaDrigar. This methodology

combines learning styles such as Accommodating,

Converging, Assimilating, and Diverging with socio-

emotional competencies (emotional awareness, emo-

tional regulation, relationship skills), fostering a per-

sonalized approach adapted to students’ demands

(Luckin et al., 2016) (Kolb and Kolb, 2005).

This paper aims to present the MannaDrigar

Methodology, developed as a strategy that combines

learning styles and socio-emotional competencies to

maximize student engagement and performance in in-

novation activities conducted in Manna_Team Boot-

camps. The methodology proposes analyzing partici-

pants’ learning styles and socio-emotional competen-

cies to organize teams with complementary proﬁles,

seeking greater engagement and synergy to foster bet-

ter performance in practical innovation activities.

As a case study, we applied the methodology in

Manna_Team Bootcamps, immersive experiences in

Exponential Technologies, with a special focus on

the Internet of Drones (IoD). During the bootcamps,

participants were divided into two groups: control

(without applying the methodology) and experimental

(with applying the methodology). These groups were

evaluated in three stages: diagnostic assessment at the

beginning of the bootcamp to measure learning styles

and socio-emotional skills; formative assessment dur-

ing the bootcamp to monitor progress in each session;

and summative assessment at the end, which mea-

sured the experimental groups’ performance through

innovation competitions. This study aims to demon-

strate the impact of personalization, based on learning

styles and socio-emotional skills, on developing tech-

nical and interpersonal skills.

Building the future involves broad access to Expo-

nential Technologies and knowledge dissemination,

making it accessible to everyone. When these tech-

nologies are used to develop soft skills and integrate

curricular content, they offer a unique opportunity for

educational innovation and disruption. This paper,

therefore, adopts the Internet of Drones as a central

technology in the bootcamp experiments, seeking to

demonstrate the potential of personalized methodolo-

gies for comprehensive student development.

2 LITERATURE REVIEW

This study builds upon two systematic literature

reviews offering a comprehensive perspective on

methodologies and technologies within the Education

5.0 framework. The ﬁrst review, "Systematic Litera-

ture Review on Instruments and Strategies for Learn-

ing Assessment in the Context of Education 5.0,"

examined methodologies and tools designed to per-

sonalize learning and integrate technical and socio-

emotional skills. Using the Parsifal tool, the initial

review retrieved 241 articles, from which 30 were

selected based on rigorous inclusion and exclusion

criteria. Among these, 12 articles underwent an in-

depth evaluation, highlighting approaches such as so-

cial interactions and collaborative platforms (25% of

the studies), survey techniques and online question-

naires (33%), as well as probing and rubric-based as-

sessments (17%). Emerging technologies, such as the

metaverse and blockchain, were explored in 8% of the

articles, while 17% recommended combining forma-

tive and summative assessments as an effective eval-

uation practice.

The second review, "Systematic Literature Review

on Artiﬁcial Intelligence in the Context of Education

5.0," explored the application of technologies such as

Artiﬁcial Intelligence (AI) and the Internet of Drones

(IoD) in educational contexts. Initially, 224 articles

were identiﬁed, with 32 pre-selected and 14 analyzed

in detail. Results revealed the growing adoption of

AI to personalize learning and provide individual-

ized feedback. Approximately 21% of the analyzed

studies examined gamiﬁcation as a motivation and

engagement strategy, while 29% highlighted recom-

mendation systems that tailor content to individual

needs, creating a dynamic and collaborative learning

environment.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

505

Both reviews also highlighted challenges, partic-

ularly in the development of socio-emotional skills,

which were identiﬁed in only 15% of the articles. This

gap emphasizes the need for further research to inte-

grate these skills into AI-based pedagogical practices,

essential for preparing students for the complex de-

mands of contemporary society.

2.1 Preliminary Studies

Annually, the Manna_Team organizes two bootcamps

during Expoingá, one of the largest agricultural fairs

in Brazil, held in Maringá. These events develop ed-

ucational methodologies and practices focused on in-

novation, emphasizing the development of technical

and socio-emotional competencies, especially for stu-

dents and teachers from public schools in the north-

west region of Paraná. The Manna Galáxias Boot-

camp hosted 65 teachers, while the Manna Agro

Bootcamp involved 119 students. In both events,

participants were challenged to use innovation tech-

niques and develop creative solutions in teams, en-

couraging collaboration and practical application of

the presented concepts.

In 2023, a study with 45 volunteers assessed

the participants’ learning styles according to Kolb’s

model, which identiﬁes four proﬁles: Accommoda-

tor, Converger, Assimilator, and Diverger. The choice

of Kolb’s model was based on the ﬁndings from

our systematic literature reviews (Section 2), which

highlighted the effectiveness of personalized teaching

methodologies based on learning styles for develop-

ing technical and socio-emotional competencies. Re-

sults showed that the Accommodator style was the

most frequent among participants, followed by Con-

verger and Diverger, indicating a general preference

for hands-on experiences.

Applying Kolb’s model in the bootcamps allowed

educational activities to align with students’ learn-

ing preferences, enhancing engagement and content

assimilation. Details on how learning styles and

socio-emotional competencies were integrated into

the study’s methodology are presented in Section 3.

The study also analyzed the organization of par-

ticipants into teams based on their learning styles and

socio-emotional competencies, investigating whether

this organization would result in better performance

compared to random team formation. It was ob-

served that participants demonstrated ﬂexibility in

their learning modes, using a secondary style that

complemented their predominant skills.

The hypothesis tested was that pre-classifying stu-

dents and teachers based on their learning styles and

socio-emotional competencies could foster a more

harmonious group dynamic, promoting academic per-

formance and personal development. To explore this

hypothesis, a correlation matrix of learning styles was

analyzed, presented in Figure 1.

Figure 1: Correlation Matrix of Learning Styles.

A high correlation (0.79) was observed between

the Accommodator and Diverger styles, and a mod-

erate correlation (0.68) between Converger and As-

similator, suggesting that predominant styles tend to

complement each other, positively inﬂuencing team

dynamics.

Based on these results and the ranking of teams at

the end of the bootcamp, this study sought to expand

the analysis of the effectiveness of pre-classifying

students, considering the development of academic

skills, collaborative abilities, and socio-emotional

skills, aligning with Manna’s mission to promote Ed-

ucation 5.0 by integrating exponential technologies

with competency development.

2.2 Study Objectives

The objective of this study is to investigate the ef-

fectiveness of the teaching, learning, and assessment

methodologies used in the IoD bootcamps promoted

by the Manna_Team. The focus is on developing par-

ticipants’ technical, collaborative, problem-solving,

and socio-emotional skills, preparing them for the

challenges of the 21st century within the context of

Education 5.0.

This study evaluates how these methodologies

contribute to the enhancement of students’ competen-

cies, examining the impact of personalized teaching

based on learning styles and socio-emotional compe-

tencies. It also seeks to measure progress in the de-

velopment of these skills throughout the bootcamp.

The hypothesis tested is:

• H1: The pre-classiﬁcation of students before the

bootcamp, based on learning styles and socio-

CSEDU 2025 - 17th International Conference on Computer Supported Education

506

emotional competencies, results in better aca-

demic performance and personal development

compared to random organization.

The research questions are:

• Q1: How do IoD bootcamps inﬂuence the devel-

opment of students’ technical skills?

• Q2: In what ways do IoD bootcamps impact par-

ticipants’ socio-emotional competencies?

• Q3: What is the effect of personalized teach-

ing, based on learning styles and socio-emotional

competencies, on students’ academic perfor-

mance?

• Q4: What are the differences in skill development

between students in the experimental group and

the control group?

• Q5: How do students perceive the effectiveness

of IoD bootcamps in their learning and personal

development?

By addressing these questions, the study aims to

provide insights into the effectiveness of the educa-

tional methodologies implemented by Manna, con-

tributing to the understanding of how the pedagogi-

cal approaches adopted in the bootcamps can prepare

students for the challenges of Education 5.0.

2.3 BootCamp Description

The objective of this study is to investigate the ef-

fectiveness of the teaching, learning, and assessment

methodologies used in the IoD bootcamps promoted

by the Manna_Team. The focus is on developing par-

ticipants’ technical, collaborative, problem-solving,

and socio-emotional skills, preparing them for the

challenges of the 21st century within the context of

Education 5.0.

This study evaluates how these methodologies

contribute to the enhancement of students’ competen-

cies, examining the impact of personalized teaching

based on learning styles and socio-emotional compe-

tencies. It also seeks to measure progress in the de-

velopment of these skills throughout the bootcamp.

The hypothesis tested is:

• H1: The pre-classiﬁcation of students before the

bootcamp, based on learning styles and socio-

emotional competencies, results in better aca-

demic performance and personal development

compared to random organization.

The research questions are:

• Q1: How do IoD bootcamps inﬂuence the devel-

opment of students’ technical skills?

• Q2: In what ways do IoD bootcamps impact par-

ticipants’ socio-emotional competencies?

• Q3: What is the effect of personalized teach-

ing, based on learning styles and socio-emotional

competencies, on students’ academic perfor-

mance?

• Q4: What are the differences in skill development

between students in the experimental group and

the control group?

• Q5: How do students perceive the effectiveness

of IoD bootcamps in their learning and personal

development?

By addressing these questions, the study aims to

provide insights into the effectiveness of the educa-

tional methodologies implemented by Manna, con-

tributing to the understanding of how the pedagogi-

cal approaches adopted in the bootcamps can prepare

students for the challenges of Education 5.0.

2.4 Bootcamp Description

This study was based on the immersive IoD boot-

camp, named "Holiday of the Beasts" ("Férias das

Feras"), held during the school vacation in Jan-

uary 2024 in the cities of Maringá, Campo Mourão,

Cianorte, and Paranavaí in Paraná, Brazil. Geared

toward public school students, the bootcamp lasted

ﬁve days, with four hours of daily activities, pro-

viding a practical and collaborative experience that

integrated the development of technical and socio-

emotional skills.

Organized by Manna, the primary objective of

the bootcamp was to introduce participants to IoD

concepts and promote essential 21st-century socio-

emotional skills, such as problem-solving, collabora-

tion, and critical thinking.

The curriculum was structured into ten modules,

covering everything from a theoretical introduction to

IoD to practical drone piloting activities and presenta-

tions of innovative projects. The topics covered were:

• Module 1. Introduction to the Internet of Drones

– Overview of IoD.

• Module 2. The Evolution of the Internet and

the Drone Era – Development of the internet and

drone integration.

• Module 3. Building a Joystick with Arduino –

Practical activity in joystick construction.

• Module 4. Drone Classiﬁcation and Applications

– Categories of drones and their applications.

• Module 5. Essential Drone Components – Study

of the main components.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

507

• Module 6. Drone Programming and Automation

– Introduction to drone programming.

• Module 7. Flight Modes and Drone Control –

Exploration of different ﬂight modes.

• Module 8. Ethics and Responsibility in Drone

Use – Reﬂection on the safe and ethical use of

drones.

• Module 9. Drone Piloting Practice – Practical

activity in drone piloting.

• Module 10 Project Presentation and Evaluation

– Presentation of the developed projects.

The "Holiday of the Beasts" bootcamp ("Férias

das Feras") provided students with an opportunity to

develop technical and socio-emotional competencies

through a practical and immersive approach. The ﬁ-

nal idea competition allowed students to demonstrate

their innovation skills and application of IoD concepts

in real-world scenarios.

This event differed from other bootcamps orga-

nized by Manna, such as those held at Expoingá, by

offering an intensive approach with an emphasis on

practical drone use and a playful, immersive experi-

ence, especially adapted for the school vacation pe-

riod. The collaborative environment and challenging

activities contributed to the enhancement of students’

skills, aligning with the principles of Education 5.0.

3 TEACHING METHODOLOGY

This section outlines the study’s methodology, cov-

ering the context of the Internet of Drones (IoD)

bootcamps conducted in four cities in Paraná, Brazil.

It describes the teaching structure based on Kolb’s

learning styles and socio-emotional competencies, the

use of the K-means algorithm for group formation,

the assessment methods (diagnostic, formative, and

summative), cluster validation techniques, and ethical

considerations.

The study took place in four cities in Paraná,

Brazil, during January 2024, involving 195 partic-

ipants: Maringá (January 8–12, 50 participants),

Cianorte (January 15–19, 78 participants), Paranavaí

(January 29–February 2, 37 participants), and Campo

Mourão (January 29–February 2, 30 participants). A

consistent methodology was applied in all case stud-

ies conducted across these locations.

The methodology combined learning styles and

socio-emotional competencies with personalization

techniques mediated by AI. Experimental groups

were formed based on Kolb’s Learning Style Inven-

tory (LSI) (Kolb and Kolb, 2005) and the Socio-

Emotional Competencies Scale (SECD) (Winsler

et al., 2014), using the K-means algorithm to clus-

ter participants into personalized and diverse groups.

Control groups were organized randomly.

The teaching methodology followed Kolb’s expe-

riential learning model (Kolb and Kolb, 2005), which

identiﬁes four learning styles:

• Converger. Applies theories to practical prob-

lems, seeking correct solutions.

• Diverger. Learns by observing and generating

ideas, viewing multiple perspectives.

• Assimilator. Organizes information logically, re-

ﬂecting on theories without immediate applica-

tion.

• Accommodator. Learns by doing, adapting to

new situations through practical experimentation.

3.1 Group Formation with K-Means

The K-means algorithm was selected for its efﬁciency

in grouping students based on quantitative character-

istics, including learning styles and socio-emotional

competencies (Hartigan et al., 1979), (Hamerly and

Elkan, 2003). Its application aimed to maximize di-

versity within each team, fostering a collaborative

and dynamic environment. (Arthur and Vassilvitskii,

2007).

Before the bootcamp, participants completed the

LSI (Kolb and Kolb, 2005) and SECD (Winsler et al.,

2014), and their scores were used for clustering. Data

normalization was conducted using z-scores to ensure

equal weighting of all variables (Felder, 2002). Four

clusters were deﬁned, aligning with Kolb’s learning

styles. The k-means++ method was employed for

initializing the clustering process, with Euclidean dis-

tance serving as the similarity metric(Arthur and Vas-

silvitskii, 2007), (Hamerly and Elkan, 2003).

Each group, whether experimental or control, was

composed of 3 to 5 students. The logistical organi-

zation allowed the bootcamps to take place simulta-

neously, ensuring that all students received the nec-

essary attention. The use of K-means promoted bal-

anced teams, where students with different proﬁles

contributed in complementary ways, optimizing tech-

nical and socio-emotional development.

The pre-tests conducted (Kolb’s Learning Style

Inventory and the Socio-Emotional Competencies

Scale) were used to form the experimental and con-

trol groups. The experimental group was organized

based on participants’ primary and secondary learn-

ing styles and socio-emotional competencies, while

the control group was randomly assigned. No ad-

ditional pre-test was conducted to compare initial

performance levels between groups, as this was not

CSEDU 2025 - 17th International Conference on Computer Supported Education

508

within the scope or objectives of the study, which fo-

cused on evaluating the impact of personalized teach-

ing methodologies during the bootcamp.

The teaching process was standardized for all

groups, with no differentiation in instructional meth-

ods. Personalization occurred exclusively in the ex-

perimental group’s team formation, where partici-

pants were grouped based on their learning styles and

socio-emotional competencies. The impact of this di-

versity, particularly during assessments and the ﬁnal

innovation competition, is analyzed and discussed in

Section 4.

3.2 Cluster Validation

To ensure the adequacy of the number of clusters, we

applied the elbow method (Ketchen and Shook, 1996)

and the silhouette index (Dudek, 2020). The elbow

method indicated that four clusters were appropriate,

aligning with Kolb’s styles. The silhouette index had

an average of 0.65, suggesting well-deﬁned clusters.

3.3 Bias Minimization and Ethical

Considerations

To guarantee impartiality and validity, teachers were

not informed of the group formations beforehand,

minimizing instructional biases. Evaluations were

blinded, ensuring that assessors were unaware of

group classiﬁcations. Standardized assessment tools

were used to ensure consistency and comparability of

results.

In ethical terms, the study followed guidelines

for research with minors, including informed consent

from parents or guardians. Data privacy was pro-

tected through anonymization, and procedures were

approved by the Research Ethics Committee.

3.4 Debates on Learning Styles

Although Kolb’s learning styles are widely used, the

literature questions their effectiveness. Studies such

as (Pashler et al., 2008) and (Kirschner, 2017) argue

that there is no robust evidence that adapting instruc-

tion to individual styles signiﬁcantly improves aca-

demic outcomes. Nevertheless, Kolb’s model was

chosen due to its relevance in the context of IoD Boot-

camps, which are intensive and hands-on. Personal-

izing instruction based on learning styles was consid-

ered effective in creating dynamic and collaborative

environments, suited to active methodologies and the

use of exponential technologies.

3.5 Availability of Research Artifacts

To promote transparency and replicability, all re-

search artifacts are available in a public repository,

including anonymized data, code used in K-means,

statistical analyses, and supplementary materials. The

repository can be accessed at:

3.6 Learning Assessment Dimensions

The assessment was structured into three dimen-

sions: diagnostic, formative, and summative, allow-

ing the tracking of participants’ development at differ-

ent stages (Black and Wiliam, 2009), (Bennett, 2011),

(Nicol and Macfarlane-Dick, 2006).

The diagnostic assessment, conducted at the be-

ginning of the bootcamp, aimed to identify students’

prior knowledge and socio-emotional competencies,

using the SECD (Winsler et al., 2014). This was es-

sential for personalizing the activities.

The formative assessment was performed at the

end of each session, with objective questions related

to the covered content, allowing continuous feedback

and immediate adjustments to activities (Halili and

Zainuddin, 2015), (Bennett, 2011).

The summative assessment took place during the

Innovation Competition, where students applied the

acquired knowledge to solve practical problems re-

lated to IoD. Project evaluation was carried out by a

jury, following a detailed rubric with criteria on orig-

inality, practical applicability, technical complexity,

and teamwork (Jones et al., 2020), (Andrade, 2005).

This combination of assessments allowed for

monitoring of students’ technical and socio-emotional

development throughout the bootcamp, ensuring a

continuous and personalized learning process.

3.7 Diagnostic Assessment

Before the bootcamp, we conducted a diagnostic as-

sessment based on the SECD (Winsler et al., 2014),

which measures skills such as emotional awareness,

emotional regulation, and relationship skills. Each

competency was rated on a Likert scale from 1 to 5,

where "Never" represents 1 point and "Always" rep-

resents 5 points.

The data collected guided group formation and

personalized activities, ensuring adequate support for

socio-emotional development (Salovey and Mayer,

1990), (Mayer, 2002), (Neubauer and Freudenthaler,

2005).

Available at: https://github.com/tmadrigar/

experimental-package-CSEDU

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

509

3.7.1 Application of the K-Means Algorithm

The K-means algorithm was applied to cluster stu-

dents based on their learning styles and socio-

emotional competencies (Hartigan et al., 1979), aim-

ing to maximize style diversity within each group

and foster a collaborative environment (Manolis et al.,

2013).

Students completed the Learning Style Inventory

(LSI) (Kolb and Kolb, 2005) and the Socio-Emotional

Competencies Scale (SECD) (Winsler et al., 2014).

The scores identiﬁed their predominant and sec-

ondary learning styles, as well as socio-emotional

skills such as self-control, empathy, collaboration,

and resilience.

The data were normalized using z-scores (Felder,

2002). We employed the k-means++ method to

strategically select initial centroids (Arthur and Vas-

silvitskii, 2007). The similarity metric used was Eu-

clidean distance (Hamerly and Elkan, 2003):

Dist(x

, c

) =

∑

d=1

i,d

− c

j,d

)

where x

represents a student’s scores, and c

is the

cluster centroid.

The K-means clustering algorithm is described as

follows:

Data: Set of students with learning style and

socio-emotional competency data

Result: Student groups clustered based on their

learning styles

Initialization:

• Select k initial centroids (or use k-means++);

• Assign each student to the nearest centroid.

while centroids are not stabilized do

for each student do

Calculate the distance between the

student and each centroid; Assign the

student to the nearest centroid;

end

for each group do

Recalculate the centroid by taking the

mean of students’ points in the group.

end

Return the formed groups.

Algorithm 1: K-Means Clustering Algorithm.

After clustering, experimental groups were

formed by considering primary and secondary learn-

ing styles, as well as socio-emotional competencies.

This approach promoted diverse teams where differ-

ent proﬁles complemented each other, maximizing

collaborative potential and aligning with the princi-

ples of Education 5.0.

3.8 Formative Assessment

Formative assessment was implemented throughout

the bootcamp to provide continuous feedback and en-

able methodological adjustments (Black and Wiliam,

2009), (Bennett, 2011).

Nine formative assessments were applied at the

end of each session, each consisting of ﬁve objec-

tive questions on the concepts covered. Responses

were collected via the Kahoot platform, which cre-

ates a real-time ranking and provides immediate feed-

back, fostering motivation and allowing adjustments

as students progressed (Wang and Hannaﬁn, 2014),

(Hamari et al., 2017).

Formative assessment is aligned with construc-

tivist learning theories, promoting the development

of metacognitive and self-regulation skills (Piaget

et al., 1952), (Vygotsky and Cole, 1978), (Nicol and

Macfarlane-Dick, 2006).

3.9 Summative Assessment

The summative assessment was conducted through an

Innovation Competition, challenging students to ap-

ply their knowledge to solve real-world problems re-

lated to IoD. Students, in teams, proposed innovative

solutions in areas such as precision agriculture, envi-

ronmental monitoring, logistics, and public safety.

We used tools like pitches and the Business Model

Canvas to structure the solutions. Teams presented

their proposals in 2-minute pitches, followed by ques-

tions from the evaluation panel.

To ensure objectivity, a detailed rubric was used

(Jones et al., 2020), (Andrade, 2005), with criteria

on Originality and Innovation (25%), Practical Appli-

cability (20%), Technical Complexity (15%), Team-

work (25%), and Presentation (15%), as shown in Ta-

ble 1.

The evaluation panel was composed of IoD ex-

perts, teachers, and industry professionals, ensuring

impartiality. Final scores were calculated based on

the established weights, resulting in the ﬁnal ranking

of each team.

4 RESULTS

This section presents the results of four case stud-

ies conducted in Maringá, Campo Mourão, Cianorte,

and Paranavaí. It includes inferential statistical analy-

ses comparing the performance of experimental and

control groups, along with a discussion of individ-

ual results for each case study. Particular attention

CSEDU 2025 - 17th International Conference on Computer Supported Education

510

Table 1: Evaluation Rubric for the Innovation Competition.

Criterion Description Weight

(%)

Level 1 (1–2

pts)

Level 2 (3–4

pts)

Level 3 (5–6

pts)

Level 4 (7–8

pts)

Level 5 (9–10

pts)

Originality

and Innova-

tion

Creativity and innovation of the

proposed solution.

25% Minimal origi-

nality

Some original-

ity

Moderately in-

novative

Innovative Highly in-

novative and

creative

Practical Ap-

plicability

Feasibility and practical applica-

bility in a real-world context.

20% Low feasibil-

ity

Some feasibil-

ity

Moderately

feasible

Feasible Highly fea-

sible and

practical

Technical

Complexity

Level of technical complexity

and appropriate use of IoD tech-

nologies.

15% Minimal com-

plexity

Some com-

plexity

Moderately

complex

Complex Highly com-

plex and

sophisticated

Teamwork Effectiveness of collaboration

among team members.

25% Minimal col-

laboration

Some collabo-

ration

Moderately

collaborative

Good collabo-

ration

Excellent col-

laboration and

dynamics

Presentation Clarity, organization, and de-

fense of the solution during the

presentation.

15% Unclear and

disorganized

Some clarity

and organiza-

tion

Moderately

clear and

organized

Clear and or-

ganized

Highly clear,

organized, and

convincing

is given to formative assessment activities and inno-

vation competitions, highlighting key ﬁndings about

the effectiveness of the personalized methodology ap-

plied during the bootcamps.

The results reﬂect the application of the method-

ology described in Section 3, which includes per-

sonalized teaching through learning styles and socio-

emotional competencies, as well as strategic group

formation.

4.1 Inferential Statistical Analyses

To evaluate the methodology’s effectiveness, inferen-

tial statistical analyses were conducted to compare the

performance of experimental and control groups. An

independent t-test revealed a statistically signiﬁcant

difference between the groups, with the experimental

group achieving higher performance. The calculated

effect size (d = 0.6) was considered moderate to high,

suggesting a relevant impact of the methodology (Co-

hen, 1988).

ANOVA was applied to assess whether sociode-

mographic variables, such as age and gender, inﬂu-

enced student performance. The analysis revealed no

signiﬁcant effects, indicating that the observed differ-

ences were primarily attributable to the pedagogical

methodology.

Conﬁdence intervals of 95% were calculated for

group means, providing greater robustness to the con-

clusions (Cumming, 2012). The analysis of effect

sizes and conﬁdence intervals reinforces that person-

alized teaching resulted in signiﬁcantly superior per-

formance in the experimental group.

4.2 Case Study 1

The ﬁrst case study, conducted in Maringá with 50

students, demonstrated that the experimental group

outperformed the control group. Figure 1 illustrates

the comparative performance in formative assess-

ments, highlighting a steeper learning curve for the

experimental group.

Figure 2: Performance of control and experimental groups

in Case Study 1.

In the innovation competition, experimental teams

achieved higher scores in originality and teamwork.

These results suggest that team formation, guided

by learning styles and socio-emotional competencies,

positively inﬂuenced student performance.

4.3 Case Study 2

The second case study, conducted in Campo Mourão

with 30 students, showed that the experimental group

performed slightly better than the control group. Fig-

ure 2 presents the comparative performance in forma-

tive assessments.

In the innovation competition, experimental teams

excelled in practical applicability and teamwork, em-

phasizing the adaptive methodology’s effectiveness in

fostering collaboration and viable solutions.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

511

Figure 3: Performance of control and experimental groups

in Case Study 2.

4.4 Case Study 3

The third case study, conducted in Cianorte with 78

students, revealed that although overall performance

was similar, the experimental group showed a clear

advantage in ﬁnal activities, suggesting better concept

retention. Figure 4 shows performance in formative

assessments.

Figure 4: Performance of control and experimental groups

in Case Study 3.

Experimental teams outperformed in applicabil-

ity and teamwork during the innovation competition,

indicating that the methodology rooted in learning

styles and socio-emotional competencies offers addi-

tional beneﬁts.

4.5 Case Study 4

The fourth case study, conducted in Paranavaí with

37 students, demonstrated a pronounced difference

between the experimental and control groups, with

the experimental group outperforming in all activities.

Figure 5 compares group performance in formative

assessments.

During the innovation competition, experimental

teams attained the highest scores across all evaluated

criteria, as shown in Table 2.

Results indicate that teams in the experimental

group excelled in all criteria, suggesting that the per-

sonalized methodology had a signiﬁcant impact on

Figure 5: Performance of control and experimental groups

in Case Study 4.

student performance.

4.6 Variability of Results

To analyze the variability of results between the

experimental and control groups, we used boxplot

graphs illustrating performance distribution in each

activity. Figure 6 presents the boxplots of formative

assessment scores for both groups across the nine ac-

tivities.

Figure 6: Performance distribution of control and experi-

mental groups.

It can be observed that experimental groups tend

to have higher medians in the activities, as well as

less data dispersion, indicating more consistent per-

formance. Control groups show greater variability in

results, with some activities presenting signiﬁcantly

lower scores.

This chart visually consolidates the positive im-

pact of the personalized methodology on student per-

formance, showing that the intervention resulted in

more stable and superior performance in the experi-

mental group.

4.7 Summary of Results

The results of the four case studies reinforce the ef-

fectiveness of the personalized methodology based

on learning styles and socio-emotional competencies.

CSEDU 2025 - 17th International Conference on Computer Supported Education

512

Table 2: Team Scores in the Innovation Competition – Case Study 4.

Team Originality Applicability Technical Complexity Teamwork Presentation

Team 1 (Exp) 9.5 9.0 8.5 9.0 9.0

Team 2 (Exp) 9.0 8.5 8.0 8.5 8.5

Team 3 (Ctrl) 8.0 7.5 7.0 7.5 7.5

Team 4 (Ctrl) 7.5 7.0 6.5 7.0 7.0

The experimental group demonstrated superior or

equivalent performance to the control group, with no-

table improvements in collaboration and practical ap-

plicability in the innovation competitions. These ﬁnd-

ings support the hypothesis that classifying students

based on learning styles and socio-emotional compe-

tencies promotes superior academic performance and

personal development.

5 INTERPRETATION OF

RESULTS

The results of the four case studies were analyzed

considering the hypothesis and research questions, fo-

cusing on the impact of the personalized methodol-

ogy—grounded in Kolb’s learning styles and socio-

emotional competencies—on students’ academic per-

formance and personal development. Personalizing

instruction through clustering based on these styles,

facilitated by the K-means algorithm, emerged as a

promising approach. Experimental groups consis-

tently outperformed control groups across various cri-

teria, as detailed in Section 4.

Student satisfaction was measured by categoriz-

ing testimonials into satisﬁed, neutral, and dissatis-

ﬁed, with scores of 5, 3, and 1, respectively. Based on

117 testimonials (99 satisﬁed, 18 neutral, 0 dissatis-

ﬁed), the average score was 4.69 on a scale of 1 to 5,

indicating a high overall satisfaction level.

5.1 Hypothesis Conﬁrmation

The central hypothesis—that grouping students based

on learning styles and socio-emotional competencies

enhances academic performance and personal devel-

opment—was conﬁrmed. Quantitative data demon-

strated that experimental groups consistently outper-

formed control groups in originality, practical appli-

cability, teamwork, and technical complexity. This

superiority was further supported by testimonials,

with 84.6% of students reporting high satisfaction.

5.2 Research Questions

Q1: How do IoD bootcamps inﬂuence the develop-

ment of students’ technical skills? The bootcamps

offered a hands-on environment emphasizing theory

application, as reﬂected in high scores for technical

complexity and practical applicability. Challenging

activities like programming and piloting encouraged

meaningful learning.

Q2: In what ways do IoD bootcamps impact par-

ticipants’ socio-emotional competencies? Exper-

imental groups exhibited stronger communication,

collaboration, and conﬂict resolution skills, under-

scoring the methodology’s positive impact on socio-

emotional development.

Q3: What is the effect of personalized teaching on

students’ academic performance? Personalized in-

struction enhanced content assimilation, as evidenced

by steeper learning curves in formative assessments.

Clear explanations and consistent support were piv-

otal to student success.

Q4: What are the differences in skill development

between experimental and control groups? Ex-

perimental groups showed superior performance in

nearly all criteria, except originality, where some con-

trol teams also excelled, suggesting that originality

may be inﬂuenced by other factors.

Q5: How do students perceive the effectiveness of

IoD bootcamps? Perception was largely positive,

with 84.6% expressing full satisfaction. Practical ac-

tivities were a highlight, though some mentioned dif-

ﬁculties with programming and limited practice time.

5.3 Variability of Results

The smaller performance difference in Case Study 3

suggests that contextual factors, such as participant

characteristics and educational environment, may in-

ﬂuence the methodology’s effectiveness, indicating a

need to adapt implementation to different contexts.

While experimental groups outperformed control

groups in practical applicability and teamwork, some

control teams excelled in originality, suggesting that

creativity and innovation may not be entirely linked

to strategic formation based on learning styles and

socio-emotional competencies.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

513

5.4 Study Limitations

Study limitations include variations in sample sizes

and contextual factors, such as student motivation and

limited practice time, which may have inﬂuenced the

results. Some students reported challenges with pro-

gramming activities, citing insufﬁcient time for prac-

tice, which could have impacted overall perceptions.

Nevertheless, teaching personalization based on

Kolb’s learning styles and socio-emotional competen-

cies was positively highlighted, enhancing academic

performance and socio-emotional development. Fu-

ture studies may explore how this personalization,

combined with active methodologies and innovative

technologies, can maximize student competency de-

velopment.

6 CONCLUSION

This study demonstrates that combining a methodol-

ogy grounded in learning styles and socio-emotional

competencies, as applied in Internet of Drones (IoD)

Bootcamps, yields positive outcomes. Personalized

instruction, paired with strategic team formation,

proves effective in enhancing technical skills and fos-

tering collaborative competencies.

Experimental groups consistently outperformed

control groups in formative assessments and innova-

tion competitions, particularly excelling in practical

applicability and teamwork. The integration of drones

and exponential technologies created an immersive

experience, enhancing the assimilation of complex

concepts and promoting creativity.

Qualitative testimonials support these ﬁndings,

highlighting students’ satisfaction with the content,

collaborative dynamics, and instructor support. An

average satisfaction score of 4.69 (on a 1 to 5 scale)

reﬂects a high level of acceptance for the methodol-

ogy.

Despite promising results, the study highlights ar-

eas for improvement, including optimizing the time

allocated for programming practices and accounting

for the speciﬁc characteristics of participants across

different educational contexts. Variability in results

suggests that the methodology’s effectiveness may

depend on contextual and individual factors.

The integration of classical and active methodolo-

gies enriched the teaching-learning process. Contin-

uous assessments conducted before, during, and af-

ter the activities enabled real-time adjustments, ensur-

ing a more personalized and relevant experience. The

use of exponential technologies, such as IoD, demon-

strated strong potential to enhance the development

of technical and socio-emotional skills, aligning with

the principles of Education 5.0.

In the landscape of educational innovation, the

Manna_Team distinguishes itself by advancing the

integration of exponential technologies into learn-

ing environments. With a commitment to inclusion,

sustainability, and personalization, Manna provides

adaptive pedagogical approaches that equip students

and teachers to tackle 21st-century challenges. Fu-

ture studies should explore how these personalized

methodologies can be reﬁned and scaled across di-

verse contexts and populations to broaden their appli-

cability.

RESEARCH

ACKNOWLEDGMENTS

We extend our gratitude to @manna_team, SOF-

TEX, and SOFTEX Campinas Nucleus, as well as

the Araucária Foundation for Supporting Scientiﬁc

and Technological Development in the State of Paraná

(FA) and the National Council for Scientiﬁc and Tech-

nological Development (CNPq) - Brazil, under pro-

cess 421548/2022-3, for their invaluable support.

REFERENCES

Andrade, H. G. (2005). Teaching with rubrics: The good,

the bad, and the ugly. College Teaching, 53(1):27–30.

Arthur, D. and Vassilvitskii, S. (2007). k-means++: The

advantages of careful seeding. In Proceedings of the

Eighteenth Annual ACM-SIAM Symposium on Dis-

crete Algorithms, pages 1027–1035. Society for In-

dustrial and Applied Mathematics.

Bennett, R. E. (2011). Formative assessment: A critical

review. Assessment in education: principles, policy &

practice, 18(1):5–25.

Black, P. and Wiliam, D. (2009). Developing the theory of

formative assessment. Educational Assessment, Eval-

uation and Accountability (formerly: Journal of per-

sonnel evaluation in education), 21:5–31.

Cohen, J. (1988). Statistical Power Analysis for the Behav-

ioral Sciences. Lawrence Erlbaum Associates, Hills-

dale, NJ, 2nd edition.

Cumming, G. (2012). Understanding The New Statistics:

Effect Sizes, Conﬁdence Intervals, and Meta-Analysis.

Routledge, New York.

Dudek, A. (2020). Silhouette index as clustering evaluation

tool. In Classiﬁcation and Data Analysis: Theory and

Applications 28, pages 19–33. Springer.

Felder, R. M. (2002). Learning and teaching styles

in engineering education. Engineering Education,

78(7):674–681.

CSEDU 2025 - 17th International Conference on Computer Supported Education

514

Floridi, L. (2014). The Onlife Manifesto: Being Human in

a Hyperconnected Era. Springer, Cham.

Ford, M. (2021). The Future of Learning: Redesigning the

21st Century Classroom. Basic Books.

Halili, S. H. and Zainuddin, Z. (2015). Flipping the class-

room: What we know and what we don’t. The on-

line Journal of Distance Education and E-learning,

3(1):15–22.

Hamari, J., Koivisto, J., and Sarsa, H. (2017). Does

gamiﬁcation work?–a literature review of empirical

studies on gamiﬁcation. Procedia computer science,

113:1101–1110.

Hamerly, G. and Elkan, C. (2003). Learning the k in k-

means. Advances in neural information processing

systems, 16.

Hartigan, J. A., Wong, M. A., et al. (1979). A k-means clus-

tering algorithm. Applied statistics, 28(1):100–108.

Hussin, A. A. (2018). Education 4.0 made simple: Ideas

for teaching. International Journal of Education and

Literacy Studies, 6(3):92–98.

Jones, I., Brown, S., and Smith, E. (2020). The impact

of rubrics on student performance and instructor feed-

back. Journal of Educational Assessment, 25(3):45–

62.

Ketchen, D. J. and Shook, C. L. (1996). The application of

cluster analysis in strategic management research: An

analysis and critique. Strategic Management Journal,

17(6):441–458.

Kirschner, P. A. (2017). Stop propagating the learning styles

myth. Computers & Education, 106:166–171.

Kolb, A. Y. and Kolb, D. A. (2005). Learning styles and

learning spaces: Enhancing experiential learning in

higher education. Academy of management learning

& education, 4(2):193–212.

Luckin, R., Holmes, W., Grifﬁths, M., and Forcier, L. B.

(2016). Intelligence Unleashed: An Argument for AI

in Education. Pearson Education.

Manolis, C., Burns, D. J., Assudani, R., and Chinta, R.

(2013). Assessing experiential learning styles: A

methodological reconstruction and validation of the

kolb learning style inventory. Learning and individ-

ual differences, 23:44–52.

Mayer, J. D. (2002). Msceit: Mayer-salovey-caruso emo-

tional intelligence test. Toronto, Canada: Multi-

Health Systems.

Neubauer, A. C. and Freudenthaler, H. H. (2005). Models

of emotional intelligence. Emotional intelligence: An

international handbook, 2005:31–50.

Nicol, D. J. and Macfarlane-Dick, D. (2006). Formative

assessment and self-regulated learning: A model and

seven principles of good feedback practice. Studies in

higher education, 31(2):199–218.

Pashler, H., McDaniel, M., Rohrer, D., and Bjork, R.

(2008). Learning styles: Concepts and evidence. Psy-

chological Science in the Public Interest, 9(3):105–

119.

Piaget, J., Cook, M., et al. (1952). The origins of intelli-

gence in children, volume 8. International Universities

Press New York.

Popenici, S. A. and Kerr, S. (2017). Exploring the impact

of artiﬁcial intelligence on teaching and learning in

higher education. Research and Practice in Technol-

ogy Enhanced Learning, 12(1):1–13.

Salmon, G. (2019). Digital education: Opportunities for

social cooperation and development. British Journal

of Educational Technology, 50(2):271–286.

Salovey, P. and Mayer, J. D. (1990). Emotional intelligence.

Imagination, cognition and personality, 9(3):185–

211.

Torres, G., Ali, R., and Goh, D. (2019). Transforming edu-

cation through technology: Challenges and possibili-

ties. International Journal of Educational Technology

in Higher Education, 16(1):1–15.

Vygotsky, L. S. and Cole, M. (1978). Mind in society: De-

velopment of higher psychological processes. Harvard

university press.

Wang, F. and Hannaﬁn, M. J. (2014). Design-based

research and technology-enhanced learning environ-

ments. Educational technology research and develop-

ment, 53(4):5–23.

Winsler, A., Kim, Y. K., and Richard, E. R. (2014). Socio-

emotional skills, behavior problems, and spanish com-

petence predict the acquisition of english among en-

glish language learners in poverty. Developmental

psychology, 50(9):2242.

World Economic Forum (2020). Schools of the future:

Deﬁning new models of education for the fourth in-

dustrial revolution. Technical report, World Economic

Forum, Geneva, Switzerland. Accessed: 2024-09-20.

Xu, Z., Chen, X., and Chen, Y. (2018). Educational robots

for 21st century skills: A review. Educational Review,

70(4):456–473.

Manna Methodology: A Novel Approach to Education 5.0 Through Learning and Socio-Emotional Assessment in IoD Bootcamps

515