Analysis of the Relationship between Self-Esteem and Depression

International Journal of Computer Science & Information Technology (IJCSIT) Vol 15, No 5, October 2023
https://doi.org/10.5121/ijcsit.2023.15502 19
ANALYSIS OF THE RELATIONSHIP BETWEEN
SELF-ESTEEM AND DEPRESSION
Rosalba Cuapa C.1
, Adair Ponce2
, Meliza Contreras3
, Mireya Tovar4
and Fernando
Zacarias F.5
1
Faculty of Architecture, Universidad Autónoma de Puebla, Puebla, México
2,3,4,5
Computer Science, Universidad Autónoma de Puebla, Puebla, México
ABSTRACT
This article presents a comprehensive and in-depth analysis of the intricate relationship that underlies self-
esteem and depression, two crucial aspects in the field of mental health. Study aims not only to scrutinize
the existing relationship between self-esteem and depression but also to do so innovatively by leveraging
the analytical and data visualization capabilities offered by two cutting-edge tools: Power BI [3], [4] and
Weka [1], [2]. These platforms, widely recognized in the realm of data science, allow for not only
exploring the relationship between these two fundamental variables but also presenting their findings in an
accessible and enlightening manner for both specialized and general audiences. Thus, our study represents
a comprehensive and multidisciplinary effort to unravel the complex relationship between self-esteem and
depression. Through the synergy between advanced data analysis carried out with Weka and the powerful
visualization provided by Power BI, the aim is not only to broaden the understanding of this crucial
relationship but also to lay the groundwork for future research and approaches in the field of mental
health. In a world where mental health takes center stage in socio-health concerns, this study aspires to
contribute valuable insights that can inform and enrich the care, treatment, and policies related to
depression and self-esteem.
KEYWORDS
Data science, Depression, Self-esteem, Weka, Power-BI
1. INTRODUCTION
Depression is a common mental disorder that affects millions of people worldwide. It is
characterized by a persistent feeling of sadness, loss of interest in activities that were once
pleasurable, changes in weight, trouble sleeping, fatigue, difficulty thinking and concentrating,
and thoughts of death or suicide. Self-esteem, on the other hand, is the way a person sees and
values themselves. It is an important dimension of mental health and can have a significant
impact on a person's well-being. People with low self-esteem often have a negative view of
themselves, feel insecure, and do not feel worthy of love or happiness. Several studies have
shown that there is a relationship between self-esteem and depression. People with low self-
esteem are more likely to develop depression than people with healthy self-esteem. This
relationship has been observed in adults, adolescents, and children. The underlying mechanisms
of the relationship between self-esteem and depression are not fully understood. However, some
researchers believe that low self-esteem can make people more vulnerable to stressors and
negative life events. Additionally, low self-esteem can make it difficult for people to cope with
life's challenges and to seek help when they need it [5].

20
This research seeks to further explore the relationship between self-esteem and depression. The
primary goal of this study is to identify the underlying mechanisms of this relationship. The
researchers will use modern tools such as Power BI and Weka to conduct complex data analysis.
The results of this research could have important implications for clinical practice. If the
researchers can identify the underlying mechanisms of the relationship between self-esteem and
depression, they could develop interventions to help people with low self-esteem reduce their risk
of developing depression. Here are some specific examples of how the results of this research
could be useful in clinical practice:
 Physicians could use the results of this research to identify people with low self-esteem who
are at an increased risk of developing depression.
 Therapists could use the results of this research to develop interventions to help people with
low self-esteem improve their self-concept and self-esteem.
 Mental health organizations could use the results of this research to develop depression
prevention programs targeted at people with low self-esteem.
 The use of modern tools such as Power BI and Weka is a strength of this research. These
tools can help to conduct complex data analysis in an efficient and effective manner. The use
of these tools could help the researchers to identify patterns and trends that might not be
visible using more traditional analysis methods.
Overall, this research is an important effort that has the potential to improve our understanding of
the relationship between self-esteem and depression. The results of this research could have
important implications for clinical practice and help to reduce the burden of depression in society.
Given that depression affects a significant proportion of the global population, understanding its
links with self-esteem stands as an objective of paramount relevance. To achieve this goal, the
study collects and utilizes data from diverse sources, enabling a comprehensive and
encompassing perspective. These data undergo rigorous analysis involving both advanced
statistical techniques and machine learning approaches, with the central support of Weka.
However, the mere acquisition and analysis of data are not sufficient to generate substantial
impact. This is where Power BI comes into play, playing a pivotal role in transforming the
analysis results into clear and effective visualizations. The platform not only allows the creation
of interactive charts and dynamic dashboards but also facilitates the effective communication of
findings to diverse audiences, including mental health professionals, researchers, policymakers,
and the public.
2. PROBLEM DESCRIPTION AND FORMULATION
To carry out this comprehensive analysis, the researchers collected data related to self-esteem and
depression from a variety of sources[6], including:
Clinical studies: The researchers reviewed the scientific literature to identify clinical studies that
investigated the relationship between self-esteem and depression. These studies provided data on
self-esteem and depression levels in groups of people with different demographic and
psychological characteristics.
Research databases: The researchers also obtained data from research databases that collect
information on people's mental health. These databases provided data on self-esteem and
depression levels in a broader population, including people who have not participated in clinical
studies.

21
The researchers used advanced data analysis and visualization techniques to analyse the collected
data. These techniques included:
 Statistical analysis: The researchers used statistical analysis to identify relationships between
self-esteem and depression levels. These analyses included correlation analysis, regression
analysis, and cluster analysis.
 Data visualization: The researchers used data visualization to represent the results of the
statistical analysis in an easy-to-understand way. The researchers created graphs, charts, and
maps to show the relationships between self-esteem and depression levels.
 The combination of Power BI and Weka provided a solid platform for exploring relationships
and patterns within the data. Power BI is a data analysis and visualization tool that allows
researchers to explore data interactively. Weka is a machine learning tool that allows
researchers to identify patterns in data.
2.1. Tools: Weka
Weka is a suite of software tools used for machine learning and data mining. The name "Weka"
comes from the acronym for "Waikato Environment for Knowledge Analysis", referring to the
University of Waikato in New Zealand, where it was originally developed by the university's
Data Mining Group. Weka provides a graphical user interface and a set of pre-implemented
machine learning algorithms that allow users to perform tasks such as classification, regression,
clustering, and association rule mining, among others. In addition to its graphical user interface,
Weka also offers a Java API (Application Programming Interface), which makes it easy to
integrate its functionality into broader applications. Weka is a powerful and versatile tool that can
be used to solve a wide range of data mining and machine learning problems. It is widely used by
researchers, academics, and practitioners in a variety of fields, including healthcare, finance, and
marketing.
Here are some of the key features of Weka [1]:
• Data Handling: Weka supports various data formats, including CSV, ARFF (Attribute-
Relation File Format), and more. It provides tools for data preprocessing, cleansing, and
transformation, making it easier to prepare data for analysis.
• Graphical User Interface (GUI): Weka offers a user-friendly GUI that allows users to interact
with the software without needing to write code. This GUI provides access to various
machine learning algorithms, preprocessing tools, and data visualization capabilities.
• Core Libraries: At the heart of Weka are its core libraries, which contain the implementation
of numerous machine learning algorithms, data preprocessing techniques, and evaluation
methods. These libraries are the foundation upon which Weka's functionality is built.
• Extensibility: Weka's architecture is highly extensible, allowing users to incorporate their
own custom machine learning algorithms, data filters, and evaluation metrics. This
extensibility is achieved through a well-defined API and plugin system.
• Cross-Validation and Evaluation: Weka includes functionality for cross-validation and model
evaluation, allowing users to assess the performance of their machine learning models. It
provides metrics like accuracy, precision, recall, F1-score, and others for model assessment.
• Integration with Other Tools: Weka can be integrated with other data analysis and
visualization tools, enhancing its capabilities, and enabling seamless workflows.
Weka is a valuable tool for anyone who needs to analyse data and extract insights. It is a
powerful and versatile tool that can be used to solve a wide range of problems.

22
2.2. Tools: Power BI
Power BI is a unified and scalable self-service business intelligence (BI) platform suitable for
large enterprises. Connect to data, visualize it, and seamlessly embed visuals into the apps you
use every day.
Power BI is a set of tools that makes knowledge available to everyone and gives us access to our
data safely and quickly, generating great benefits for us and for our company. It is a predictive,
intelligent, and highly supportive system, capable of translating data (simple or complex) into
graphs, panels or reports due to its qualities such as the graphical capacity of presenting
information, or the integration of Power Query: the search engine extract, transform, and load
(ETL) software included in Excel.
Here are some of the key features of Power BI:
• Easy handling: you will not need to have advanced knowledge to create your own
dashboards and reports. Simply build, edit, publish and share to collaborate with whoever
you need.
• Integration capacity: Take advantage of the integration capacity of data sources as different
as those that Salesforce, MailChimp, SAP BW, SAP HANA, MySQL, SQL Server, Teradata,
Oracle, Google, Twitter, Facebook, Dynamics 365 can provide you.
• Attractive graphics: Visualize your data with amazing, animated infographics or if you
prefer something more traditional, use the usual graphics.
• Seguridad: Con power BI podrás establecer perfiles o roles de acceso a la información para
cada miembro de tu empresa.
• Información en tiempo real: Si es necesario podrás programar la descarga de datos para tus
KPI´s tanto como necesites y acceder a información actualizada.
2.3. Comparative Analysis of Business Intelligence platforms
Based on comparative analyzes presented on BI tools such as: Joaquim Lapa, et.al. [12], Nuno
Leite, et.al. [13], etc. and considering the criterion used in Gartner [14] we can paraphrase it as
follows: The analysis of BI platforms should be considered from three broad categories:
integration, information delivery and analysis. These criteria allow organizations to have clear
classification and measurement systems to support decision making. In addition, we must
consider the facilities and benefits that these tools provide us.
3. EXPERIMENTAL STUDY
the data were obtained from multiple sources revealed a notable negative correlation between
self-esteem and depression. Individuals with low self-esteem exhibited a higher incidence of
depressive symptoms compared to their counterparts with higher self-esteem. Additionally, a
mediating role played by self-esteem was observed in the relationship between certain risk
factors and depression. This finding underscores the importance of self-esteem as an influential
variable in the development of depression.
3.1. Database Design
The database was designed for a type of archive native for weka, the type of this archive is
ARFF:

23
• @relation: This command defines the name or title of the dataset. It specifies the dataset's
overall name or a brief description.
• @attribute: This command defines the attributes or columns of the dataset. It specifies the
names of the attributes, their types (such as numeric, string, or nominal), and their possible
values or value ranges.
• @data: This command marks the beginning of the data section of the ARFF file. In this
section, actual data instances are listed, with each instance's values corresponding to the
attributes defined earlier.
The database was recovered from the official page of "The INEGI" (Instituto Nacional de
Estadística y Geografía) is the Mexican National Institute of Statistics and Geography. It is a
government agency responsible for collecting, analyzing, and disseminating statistical
information and geospatial data for Mexico. INEGI plays a crucial role in providing accurate and
up-to-date information on various aspects of the country, including population, economics,
geography, and demographics.
3.2. First Source of Information
Methods: a cross-sectional study was carried out in 1,104 freshman students, 40.3% men and
59.7% women, at a public university in Mexico City. The 20-item depression scale (CES-D) and
Food Frequency Questionnaire were applied to measure depressive symptoms and food
consumption. Logistic regression analysis was carried out for food consumption frequency and
CES-D depression score grouped in quartiles [7].
Results: the prevalence of depression symptoms was 18.2% in men and 27.5% in women (p <
0.001). A considerable proportion of the students reported poor eating habits: consumption of
fried food (30.3%), sweetened drinks (49.0%) and sugary food (51.8%) 2-7 times/week; and less
than half the students practiced vigorous physical activity (39.7%). In women, a higher
depression score was associated with a higher frequency of consumption of fast food (OR = 2.08,
p = 0.018), fried food (OR = 1.92, p = 0.01) and sugary food (OR = 2.16, p = 0.001), and a lower
frequency of physical exercise (< 75 min/week; OR = 1.80, p = 0.017). In men, no association
was observed between depression score and food consumption variables. An association was
observed between depression and low exercise frequency (OR = 2.22, p = 0.006).
3.3. Second Source of Information
Data collected by the National Institute of Statistics and Geography (INEGI) of Mexico also
provided crucial information for this analysis. The numbers and statistics provided by this entity
enriched the understanding of the prevalence of depression in the population and how it may be
related to self-esteem.
Questions asked in the survey:
1. I couldn't shake the sadness.
2. Never or rarely: I never felt this way.
3. A few or sometimes: I felt this way a few times.
4. A considerable number of times: I felt this way a lot of times.
5. All the time or most of the time: I felt this way all the time or most of the time.
6. Difficulty concentrating.
7. Never or rarely: I never had difficulty concentrating.
8. A few or sometimes: I had difficulty concentrating a few times.

24
9. A considerable number of times: I had difficulty concentrating a lot of times.
10. All the time or most of the time: I had difficulty concentrating all the time or most of the
time.
11. Feeling depressed.
12. Never or rarely: I never felt depressed.
13. A few or sometimes: I felt depressed a few times.
14. A considerable number of times: I felt depressed a lot of times.
15. All the time or most of the time: I felt depressed all the time or most of the time.
16. Everything felt like an effort.
17. Never or rarely: I never felt like everything was an effort.
18. A few or sometimes: I felt like everything was an effort a few times.
19. A considerable number of times: I felt like everything was an effort a lot of times.
20. All the time or most of the time: I felt like everything was an effort all the time or most of
the time.
21. Didn't sleep well.
22. Never or rarely: I always slept well.
23. A few or sometimes: I had trouble sleeping a few times.
24. A considerable number of times: I had trouble sleeping a lot of times.
25. All the time or most of the time: I had trouble sleeping all the time or most of the time.
26. Enjoying life.
27. Never or rarely: I always enjoyed life.
28. A few or sometimes: I enjoyed life a few times.
29. A considerable number of times: I enjoyed life a lot of times.
30. All the time or most of the time: I enjoyed life all the time or most of the time.
31. Feeling sad.
32. Never or rarely: I never felt sad.
33. A few or sometimes: I felt sad a few times.
34. A considerable number of times: I felt sad a lot of times.
35. All the time or most of the time: I felt sad all the time or most of the time.
3.4. Third Source of Information
Another important study, conducted by Sánchez-Rojas et al. (2022), examined children and
adolescents with and without overweight or obesity. The study highlights the correlation between
self-image, self-esteem, and depression. The results show that self-image and self-esteem
influence the presence of depressive symptoms in these groups. This study provides valuable
insights into how body image and self-evaluation can influence mental health [9], [10].
Methods: A cross-sectional comparative study of 295 children: 116 children with overweight or
obesity (group 1) and 179 without obesity (group 2). Body mass index, schooling, academic
achievement, school conflicts, socioeconomic status, self-image (current, desired), satisfaction,
self-esteem, and presence of depression were recorded. Descriptive statistics, Spearman's rho,
and Mann-Whitney U were used; p ≤ 0.05 was considered significant.
Results: In group 1, 53.4% of the children perceived themselves as normal weight and 77.6% had
a desired self-image of normal weight; 67.2% wanted to be thinner; 53.4% had high self-esteem;
75.9% were observed without depression. In group 2, 79.3% had a current self-image of normal
weight and the desired self-image in 85.5% was normal weight; 35.2% wanted to be thinner; self-
esteem was high in 49.7% and 77.1% did not present depression. Significant correlations were
identified in self-esteem-depression (r = 0.228) and self-esteem-socioeconomic status (r = 0.130).
The data obtained from multiple sources revealed a notable negative correlation between self-
esteem and depression. Individuals with low self-esteem exhibited a higher incidence of

25
depressive symptoms compared to their counterparts with higher self-esteem. Additionally, a
mediating role played by self-esteem was observed in the relationship between certain risk factors
and depression. This finding underscores the importance of self-esteem as an influential variable
in the development of depression.
Table 1. Data Within Weka, showing the minimum, maximum, mean, and standard deviation
table at the top.
Figure 1. Attribute “Mujer”
Table 2. Data Within Weka, showing the minimum, maximum, mean, and standard deviation
table at the top.

26
Figure 2. Attribute “Hombre”
Figure 3. File ARFF
Average life satisfaction (1) of the literate population aged 18 and over (2), by symptoms of
depression (3) according to sex and age groups (18 to 29 years old).
(1) To measure the variable, a card with a scale of 0 to 10 is used, in which 0 means "totally
dissatisfied" and 10 "totally satisfied".
(2) The ENBIARE interview process requires Spanish reading and writing skills on the part of the
informant.
(3) To capture this topic, a card with a textual and numerical scale is used, presenting the numbers
from 0 to 3 and the meaning of each one: 0 = rarely or never (less than 1 day), 1 = few or
sometimes (1-2 days), 2 = a considerable number of times (3-4 days), and 3 = all the time or most
of the time (5-7 days).
The dependent variable in this case is named "Value", it depends on the variable "Man" and
"Woman", in addition to the variable "SymptomsDepresion" and "Frequency", of these depends
"Man" and "Woman".

27
The INEGI data collection methodology involves a variety of sources and techniques. These
include household surveys, establishment surveys, population and economic censuses, among
others. Each of these sources provides a unique perspective on the reality of Mexico and its
inhabitants. Through the collection of data in diverse contexts, INEGI seeks to guarantee the
accuracy and representativeness of its statistical information. The data sets generated by INEGI
have a significant impact on decision-making at the government, economic, and social levels. The
results of surveys and statistical analyzes provide crucial information for understanding
demographic, economic, and social trends in Mexico. These data are used to inform public
policies, guide investments, evaluate programs and projects, and make informed business
decisions. That is, these data that it provides us in themselves are the DataSets that INEGI
generates, they are a source to obtain DataSets, in this case they are not processed because it
shares them in a way that is difficult to analyze, however, in this case what was done was to adapt
them to an ARFF file to be able to be analyzed with the Weka tool. The importance of INEGI's
data sets also lies in the transparency and access to the information they provide. The availability
of detailed and up-to-date statistical data allows researchers, academics, and citizens in general to
analyze the information for themselves and generate new knowledge from it. This contributes to a
greater understanding of the problems and challenges facing Mexico and promotes a dialogue
based on solid data.
4. CONCLUSIONS
This comprehensive analysis underscores the need to consider self-esteem as a critical element in
depression research. Self-esteem not only appears to play an essential role in the onset and
development of depression, but it can also influence the effectiveness of therapeutic
interventions. The results highlight the importance of taking self-esteem into account in the
design of depression prevention and treatment strategies. It is essential to recognize that this
analysis has inherent limitations. The data were collected from various sources with different
methodologies, which can affect the quality and consistency of the results. In addition, the
correlational nature of the data prevents definitive causal relationships from being established.
Longitudinal and controlled research is suggested to delve into the relationship between self-
esteem and depression.
5. FUTURE DIRECTIONS
This report provides a solid perspective on the relationship between self-esteem and depression
and highlights the usefulness of visualization tools such as Power BI and Weka to better
understand data patterns. However, to achieve a deeper and more accurate understanding of this
relationship, a continuous focus on further research and the improvement of data analysis
techniques is required. A full understanding of how self-esteem impacts depression could lead to
significant advances in the diagnosis and treatment of this disorder [11].
ACKNOWLEDGEMENTS
This research was supported/partially supported by Vicerrectoria de Investigación y Estudios de
Posgrado with Proyect ID: 00082. We thank our colleague from Faculty of Architecture at BUAP
who provided insight and expertise that greatly assisted the research. We thank to Adair Ponce,
student of computer science for assistance with technique and methodology. Finally, we thank
Meliza and Mireya for their understanding and computational expertise in data science.

28
REFERENCES
[1] ¿Qué es Weka y qué tiene que ver con Big Data? (2020, julio 9). Agenciab12.mx.
https://agenciab12.mx/noticia/que-es-weka-que-tiene-que-ver-big-data.
[2] Donatella Merlini, Martina Rossini, Text categorization with WEKA: A survey,Machine Learning
with Applications,Volume 4,2021,100033,ISSN 2666-8270,
https://doi.org/10.1016/j.mlwa.2021.100033.
(https://www.sciencedirect.com/science/article/pii/S2666827021000141)
[3] Gurpreet Singh, Ankul Kumar, Jaspreet Singh, Jagdeep Kaur. (2023) Data Visualization for
Developing Effective Performance Dashboard with Power BI. 2023 International Conference on
Innovative Data Communication Technologies and Application (ICIDCA), pages 968-973.
[4] Cree informes sofisticados y comparta conocimientos que impulsen los resultados. (s/f).
Microsoft.com. Recuperado el 24 de agosto de 2023, de https://powerbi.microsoft.com/es-
mx/landing/free-account/
[5] Arévalo García, E., Castillo-Jimenez, D. A., Cepeda, I., López Pacheco, J., & Pacheco López, R.
(2019). Anxiety and depression in university students: relationship with academic performance.
Interdisciplinary Journal of Epidemiology and Public Health, 2(1), e–022.
https://doi.org/10.18041/2665-427X/ijeph.1.5342 (Original work published July 15, 2020)1.
[6] Mishra P, Pandey CM, Singh U, Gupta A, Sahu C, Keshri A. Descriptive statistics and normality
tests for statistical data. Ann Card Anaesth. 2019 Jan-Mar;22(1):67-72. doi:
10.4103/aca.ACA_157_18. PMID: 30648682; PMCID: PMC6350423.
[7] Sánchez-Rojas, Alma A., García-Galicia, Arturo, Vázquez-Cruz, Eduardo, Montiel-Jarquín, Álvaro
J., & Aréchiga-Santamaría, Alejandra. (2022). Autoimagen, autoestima y depresión en escolares y
adolescentes con y sin obesidad. Gaceta médica de México, 158(3), 124-129. Epub 28 de
septiembre de 2022. https://doi.org/10.24875/gmm.21000817.
[8] Lazarevich I, Irigoyen Camacho ME, Velázquez-Alva MC, Flores NL, Nájera Medina O, Zepeda
Zepeda MA. Depression and food consumption in Mexican college students. Nutr Hosp. 2018 May
10;35(3):620-626. English. doi: 10.20960/nh.1500. PMID: 29974771.
https://pubmed.ncbi.nlm.nih.gov/29974771/
[9] Instituto Nacional de Estadística y Geografía (INEGI). (2022). Buscador INEGI.
https://www.inegi.org.mx/app/buscador/default.html?q=Depresión.
[10] Instituto Nacional de Estadística y Geografía (INEGI). (s/f). Quiénes somos. Recuperado el 19 de
agosto de 2023, de https://www.inegi.org.mx/inegi/quienes_somos.html.
[11] Fan, C., Chen, M., Wang, X., Wang, J., & Huang, B. (2021). A review on data preprocessing
techniques toward efficient and reliable knowledge discovery from building operational data.
Frontiers in Energy Research, 9, 652801
[12] Joaquim Lapa, Jorge Bernardino and Ana Figueiredo. A comparative analysis of open-source
business intelligence platforms. Conference ISDOC: Information Systems and Design of
Communication 2014. DOI: 10.1145/2618168.2618182
[13] Nuno Leite, I. Pedrosa, Jorge Bernardino. Comparative evaluation of open source business
intelligence platforms for SME Published 2018 13th Iberian Conference on Information Systems
and Technologies (CISTI).
[14] Group, Gartner. Magic Quadrant for Business Intelligence andAnalytics Platforms.
http://www.gartner.com. [Online] 2013.
AUTHORS
Fernando Zacarias-Flores is researcher and professor of Computer Science at the
Autonomous University of Puebla. He is a researcher in practical and theoretical
Computer Science and Mobile Technologies and has conducted R&D projects in this
area since 2000. Results from these projects have been reported in more than 80
national and international publications.

Analysis of the Relationship between Self-Esteem and Depression

More Related Content

Similar to Analysis of the Relationship between Self-Esteem and Depression (20)

Recently uploaded (20)

Analysis of the Relationship between Self-Esteem and Depression