CHAPTER 1

THE PROBLEM AND THE BACKGROUND

This chapter contains the background of the study, statement of the problem, significance of the
study, scope, limitation and delimitation.

Background of the Study

Society has changed dramatically from a traditionally driven society toward a modern knowledge-
based society that is fueled by innovation and creativity. Artificial Intelligence (AI) has become a widely
discussed topic in the present age as technology continues to advance rapidly. It is now acknowledged that
artificial intelligence is the technology of the future on a national and worldwide scale (Krstić, Aleksić,
and Krstić, 2022). As stated by Jain & Jain (2019), the technical revolution in AI has occurred in numerous
regions across the globe in the last few decades. It seeks to make computers learn by simulating human
intellect.

The study emphasizes the importance of examining how Artificial Intelligence (AI) affects
criminology students at Palawan State University's main campus regarding its potential influence on their
independent thinking. There is concern that students may become unduly dependent on these tools as Al
becomes more and more common in the classroom, which could hinder the development of critical
thinking and problem-solving abilities. In order to pinpoint particular difficulties and possibilities in their
educational setting, the study focuses on criminology students at Palawan State University to identify
specific challenges and opportunities within their learning environment. In addition to understanding the
effects of Al, this study attempts to create ways that enable pupils to succeed in the face of these
technological improvements.
Artificial intelligence (AI) is revolutionizing education by enabling more individualized learning
experiences, including personalized learning experiences through platforms like ChatGPT, OpenAI,
Grammarly, Copy.ai, and Quillbot. By interacting with students as though they were human teachers, these
Al tools increase productivity and efficacy. However, there are concerns about possible negative effects
including impairing critical thinking and decreasing interpersonal communication (Greene-Harper, 2023).
The purpose of this study is to investigate the impact of artificial intelligence on the academic
performance of criminology students in Palawan State university. With the rapid advancements in
technology, artificial intelligence has become increasingly prevalent in various industries, including
education. However, there is a lack of research on the effects of Al specifically in the academic realm. This
study seeks to fill that gap by examining how the use of Al tools and resources affects the academic
achievement of the students as well as to provide a plan of action on how to minimize the potential
negative impacts of artificial intelligence on students.
Statement of the Problem

This study aims to determine the impact of artificial intelligence (AI) on the academic
performance of the criminology students at Palawan State University.

Specifically, this study seeks to answer the following questions:

1. What is the demographic profile of the criminology students in terms of;
1.a Sex
1.b Age
1.c Year Level
2. What extent do criminology students at Palawan State University use artificial intelligence (AI)
tools and resources for their academic work?
2.a Frequency
2.b Purpose
3. What is the level of academic performance among criminology students at Palawan State
University who have access and utilize artificial intelligence in terms of;
3.a Writing skills
3.b Critical thinking skills
4. Is there a significant relationship between the students' use of Al academic tools and their
academic performance?

Significance of the Study

The study aims to determine the Impact of Artificial Intelligence to the Academic Performance of
Criminology Students at Palawan State University, contributing important insights to the global
conversation on AI in education. This research could provide valuable guidance and support in the
following areas:
Students. The result might provide students with knowledge on the benefits and challenges
associated with Al tools and how to effectively and wisely integrate these technologies into their studies
potentially improving their overall academic performance and learning outcomes.
School Administration. The result might provide a guide for administrators to design policies and
support systems that promote responsible and effective use of AI tools, ensuring students get a high-quality
education that keeps up with modern technology.
Teachers. The result might benefit the teachers understanding how AI tools affect students'
performance and raise awareness about potential over-reliance, making sure students gain the benefits of
technology while still developing important thinking and analytical skills.
Future Researchers. The result could provide a basis for future research, helping researchers
investigate specific aspects of AI use in education.
Scope and Delimitation
This study focuses on assessing the impact of artificial intelligence on the academic performance
of criminology students at Palawan State University Main Campus as well as to identify if there is a
significant impact on their academic performance. To be able to gather data, the criminology students at
Palawan State University will be the main respondents of this study. The respondents will come from
criminology students of all levels (1 st year to 4th year) of College of Criminal Justice Education in Palawan
State University Main Campus. This study will use questionnaires that will be directly answered by their
respondents.
It delimits that the study will not include other students from other colleges in Palawan State
University Main Campus. Also, this study only looks at how artificial intelligence affects students' grades
and learning, not how it might affect other things in our community.

Definition of Terms
To have a better and through understanding of the study, the following terms have been conceptually
and operationally defined.
Artificial Intelligence. It refers to the emulation of human intelligence function by machines or
computer system (Coursera 2023). In this study, artificial intelligence refers to the technology that helps
students by providing smart tools and system designed to enhance their learning experience.
Academic Performance. It refers to the outcome of students’ efforts in in achieving goals across
various academic subjects. In this study, academic performance specifically refers to the grades obtained
by students in their subjects where artificial intelligence is integrated
Critical Thinking. It is the ability to interpret, evaluate and analyze facts and information that are
available to form a judgement. In the study, critical thinking specifically relates to the cognitive abilities
demonstrated by the students at San Luis Senior High School when using artificial intelligence tools and
resources.
Frequency. It refers to the number of times an event occurs within a specific period. In this study,
frequency refers to the regularity of engagement with AI-based learning tools and resources by students.
Purpose. It means specific life goals or just a feeling of motivation to go about your daily to-dos
(Hirsch, 2022). In this study, purpose is defined as the perceived educational goals and intentions of
students when utilizing AI tools in their research activities.
Writing Skills. These are the specific abilities that help writers put their thoughts into words in a
meaningful form (Harmer, 2017). In this study, writing skills specifically refer to the proficiency
demonstrated by students when composing written assignments that involve the use of AI-generated
content or assistance.
 CHAPTER II

REVIEW OF RELATED LITERATURE AND STUDIES

This chapter contains related literature and studies that are significant to this study and were gathered from
local and foreign sources. This also includes the theoretical framework, the research paradigm, the research
hypothesis, and the definition of terms.

Related Foreign Literature

2.1. Artificial intelligence overview

AI or machine intelligence is an area of computer science where machines are programmed with
the ability to perform intelligent tasks that are usually undertaken by humans (Dwivedi et al., 2023c; Tsang
et al., 2020; Ali et al., 2023; Pan and Nishant, 2023). Computers and machines use AI techniques to
understand, analyze, and learn from data through specifically designed algorithms (Sasubilli et al.,
2020; Richey Jr et al., 2023). For example, with modern AI technologies, cameras can automatically
recognize faces, while computers can translate between one language to another (Sasubilli et al., 2020). AI
has been founded as an academic discipline since the 1950s and since then, it has been significantly
researched in areas such as NLP, learning, reasoning and various knowledge domains.
More recently, AI has been transformed with the expansion of its research beyond computer
science, with recent developments drawing from broad areas such as psychology, linguistics, and
philosophy (Ali et al., 2023). Consequently, AI has been applied in various areas such as education, e-
commerce, robotics, navigation, healthcare, agriculture, military, marketing and in gaming consoles. More
specifically, widely adopted AI applications include search engines such as Google, recommender
systems such as Netflix, self-driving cars such as Tesla, and human speech recognition systems such as Siri
and Alexa. In general, AI methods can be broadly categorized in these areas as machine
learning (Bernardini et al., 2021), robotics NLP (Murray et al., 2019), computer visioning (Jahan and
Tripathi, 2021), and big data (Hossen and Karmoker, 2020).
Classification and clustering are two major techniques used in AI machine learning. Both algorithms use
data such as numbers, text, images and videos, as input (Jahan and Tripathi, 2021). Classification
algorithms (such as neural networks, decision trees and Bayesian networks) use huge amounts of data as
training datasets. There are two types of classification algorithms: supervised and unsupervised
learning (Uddin et al., 2019). Supervised learning uses labeled data vectors during training; by
contrast, unsupervised learning algorithms do not use labels. Both methods use class labels during the
testing phase. In machine learning, clustering algorithms are used for unsupervised learning and do not
need any class label data whereas prediction algorithms are trained using historical data to
develop forecasting models (Libbrecht and Noble, 2015). Several algorithms are used in classification,
clustering and prediction (Elbasi et al., 2021):
AI and its sub-areas such as robotics, Internet of Things (IoTs), and machine learning can have significant
impacts on society. AI technology can improve human life quality, making life easier, safer and more
productive (Chaturvedi et al., 2023; Malik et al., 2021; Hradecky et al., 2022). There are
several application areas of AI that make human life easier such as face recognition for security,
automation for industry, NLP for translation, and robotics for homes (Herath and Mittal, 2022). AI has
transformed our society to move into the Industry 4.0 revolution due to the IoTs, cloud computing,
robotics, cyber physical systems and machine to machine communication (Votto et al., 2021). When used
effectively, the smart automation and interconnectivity can allow people to save time, manage work
flexibility and increase collaborations (Ahsan and Siddique, 2022).

2.2. Generative AI types

Generative AI can be defined as a technology that (1) leverages deep learning models to (2) generate
human-like content (e.g., images, words) in response to (3) complex and varied prompts (e.g., languages,
instructions, questions) (Lim et al., 2023). Generative AI models are AI platforms that generate a variety of
outputs based on massive training datasets, neural networks and deep learning architecture, and prompts
from users (Nirala et al., 2022). Depending on the type of generative AI model, various models can
possibly generate images, translate text into image outputs and vice-versa, synthesize speech and audio,
create original video content, and generate synthetic data (Porkodi et al., 2022). Although there are many
different subsets and new formats of generative AI models emerging, the two primary designs
are: Generative adversarial networks (GANs) and AI called transformer-based models. With generative AI,
the components of the model include two different neural networks: the generator and the discriminator.
The generator compiles content based on user inputs and training data while the discriminator model
evaluates generated content against “real” examples to determine which output is real or accurate (Gonog
and Zhou, 2019). With the transformer-based model, encoders and/or decoders are built into the platform
to decode the tokens or blocks of content that have been segmented based on user inputs (Li et al., 2022).
The primary difference between generative and discriminative AI models is that the former can create new
content and outputs based on their training (Qadir, 2023). Discriminative modeling, on the other hand, is
primarily used to classify existing data through supervised learning (Van Engelen and Hoos, 2020). As an
example, a protein classification tool would operate on a discriminative model, while a protein generator
would run on a generative AI model. Generative models are designed to create something new while
predictive AI models are set up to make predictions based on data that already exists. Continuing with our
example above, a tool that predicts the next segment of amino acids in a protein molecule would work
through a predictive AI model while a protein generator requires a generative AI model approach ( Thomas
et al., 2023).
In addition, another AI model Variational Autoencoders (VAE), is used for text and audio generation; VAE
is a generative model that encodes data into an embedded space and then decodes it to reconstruct the
original content. VAE models use distinct probabilistic combinations of input data to generate new content
(Yadav et al., 2021). Similarly, Autoregressive Models (ARM) generate one-unit element at a time,
deriving cues from the earlier generated element (Bai et al., 2021). This regressive one-unit at a time
function aids in creating contextual and yet coherent content (GPT is one of these types). Recurrent Neural
Networks (RNN) is also an AI model that processes sequential data by predicting the next unit element
from the previous element; unlike ARM, they are neural networks and lack the potential to generate long
sequences of data; functional improvements are currently being developed to overcome the limitations of
RNN (Chen et al., 2019). Transformer-based models have raised wider acceptance since unlike RNN, they
can handle long output sequences of data by creating elaborate, coherent and contextual content. Flow-
based generative models have the capacity to portray the data distribution by inverting the metamorphosis
between the prompt and generated output. These models aid in generating data as well as density
estimation of the data generated.
It is very important to acknowledge the power of generative AI with its related concepts. In line with our
definition, it is worth noting that generative AI has the unique ability to not only provide a response but
also generate the content in that response, going beyond the human-like interactions in conversational AI
(Lim et al., 2022). In addition, generative AI can create new responses beyond its explicit programming,
whereas conversational AI typically relies on predefined responses. However, not all generative AI is
conversational, and not all conversational AI lacks the ability to generate content (Lim et al., 2023).
Augmented AI models, such as ChatGPT, combine both generative and conversational AI to enhance their
capabilities (Dwivedi et al., 2023a). Additional background details about ChatGPT are discussed next.

2.3. Generative AI models and background to ChatGPT

Generative AI is a distinct class of AI and an incredibly powerful technology that has been popularized by
ChatGPT (Lim et al., 2023). Open Artificial Intelligence (OpenAI) is the source of ChatGPT, which is a
format of a large language model (Abdullah et al., 2022), which is structured to synthesize human-like text
based on an array of input commands. ChatGPT is useful for a spectrum of Natural Language Processing
assignments such as script generation, comprehension of those scripts and conversations, and their
translation (Kirmani, 2023). The introduction of ChatGPT achieved one million users within a week of its
release on November 30, 2022 (Altman, 2022; Mollman, 2022; Hu, 2023), shocking users with its degree
of sophistication and human-like intelligence exhibited on specific prompts and commands. This gathered
attention of social media, news and research-oriented platforms like Nature (Stokel-Walker, 2022; Metz,
2022). The technology can process multiple, complex and comprehensive tasks including drafting articles
(Stokel-Walker, 2022), summarize content, sign-scripting to address distinct criteria and perform
specialized functions such as drafting and de-bugging computer code. This has led to many eclectic
responses from experts in academic and educational settings as the application has not only been disruptive
but revolutionary in terms of its scholarly and pedagogical capacity (Williams, 2023). At its most basic, the
program has dramatically increased user ability to create knowledge and the means by which it is accessed
(Lucy and Bamman, 2021). As the authors discuss below however, opponents are vocal that the knowledge
produced creates many ethical dilemmas thus compromising human ingenuity the quality of the teaching
and learning process (Williams, 2023).
As discussed, the presence of diverse Generative AI aids in the creation of image, audio, text and visual
content; this has increased the usefulness of AI to academic as well practitioners of diverse domains. That
is, diverse AI tools with their embedded functional and operational utility have the latent ability to create
credible content within a fraction of the time taken in traditional learning models ( Kar et al., 2023). They
can also be customized to initiate sequential prompts. While current Generative AI aids may look
revolutionary, their origin can be attributed to the 1960s when Chatbots were invented. Mostly, AI models
became invasive around 2014 with the inception of GANs, which is a typology for machine learning
algorithms (Behrad and Abadeh, 2022). This breakthrough technology and its application in diverse fields
has led to its application and adoption in a variety of technical, intellectual, business and operational
contexts e.g., the movie industry and in academic writing inter alia. Transformer technology which is
based on machine learning has revolutionized learning by enabling the training of AI models to embed
colossal content without the necessity to label the content in advance. These transformer aid models create
a nexus of prompts across, pages, data-sets, books and input chapters, compared to their predecessors who
were restricted to searching for sentences and words. Thus, AI models have the ability to revolutionize
nascent fields like Biotechnology for instance since they can create connections across bio-codes, bio-
chemicals, proteins and DNA strands. Large Language Models, of which ChatGPT is a type, have
unearthed a plethora of generative AI models that can create images, audio and visual content from trivial
prompts (Kumar, 2023).
ChatGPT materializes a transformer architecture, a computer-enabled neural nexus that accentuates
the NLP abilities of an artificial system. This architecture is connected to a large input data-set and is
integrated to devise transcripts based on its data source though creative blending (Baidoo-Anu and Ansah,
2023). The input command of the application processes and furnishes an output for each unit of transcript
as one-unit time. The whole output is attached to the preceding unit-output in the string in connection with
the linkage prompt/command that was rendered. The application materializes attention machinery to focus
on the maximum segment of the input to produce an output that is intelligent, comprehensive and
customized to the input received. ChatGPT can be seasoned and made comprehensive on specialized
actions like contextual dialogue-generation or query-resolution systems by offering an additional command
of task-specific input by improvising it for the specific NLP application. The mechanism can be configured
for distinct dialects and vernaculars by customizing the input datasets or by prototyping it with specific
language computer codes (Kasneci et al., 2023).
The COVID-19 epidemic has been disruptive highly disruptive in educational setting in relation to how
educational content was delivered with most major academic institutions transitioning to electronic,
remote, and online learning, to conform to social distancing guidelines (Chatzipanagiotou and Katsarou,
2023). Significant change to this extent was a disruptive shift to digital and online pedagogy, as institutions
and the international community mandated quick adoption of the technology-enabled teaching-learning
process in the face of increasing adversity (Coghlan et al., 2021; Henderson et al., 2022). The pandemic
had effectively curtailed the face-to-face learning system meaning a major transition and unprecedented
embrace of technology were required in the teaching-learning space. This has included the acceptance of
online virtual conversational platforms such as Google Classroom, Teams, Zoom, and other different video
tools (e.g. online conferences) which extended to other pedagogical tools like e-books, videos, and
interactive activities (Chatzipanagiotou and Katsarou, 2023). Use of sophisticated learning management
systems like Moodle, Google Suite, has further empowered the teaching-learning fraternity with new
teaching aids. Thus, the advent of different e-learning platforms has revolutionized the delivery of
education which has had to become more agile to invite the participation of students including remotely.
Moreover, the global pandemic invited the need for more self-dependence and asynchronous learning
systems. Here, learners now require substantial independence in how they learn and the speed by which
they learn. Artificial intelligence and generative models such as ChatGPT accounts for the new
requirements and convenience of learning and at least in theory, can embody a learner's socio-cultural
background. However, this AI transition has magnified many pedagogical issues related to quality
including the digital divide between those who have access to sophisticated technology and those who
don't (Cain, 2023). Generative AI models have brought to the surface other drawbacks as well such as
restricted interaction, a dearth of academic readiness, and issues of ethics and poor accountability ( Baidoo-
Anu and Ansah, 2023; Nguyen et al., 2023; Yan et al., 2023; Stahl and Eke, 2024). Holistically, the
pandemic catalyzed the adoption of technology in education while underscoring problems of learner-
inclusivity and accessibility (Chatzipanagiotou and Katsarou, 2023). These facts have brought a greater
focus on global education systems. For instance, the education system is invested with the responsibility to
perennially adapt, evolve and bridge the gap between the stakeholder (student, teachers, and parents) needs
during challenging times (Schiff, 2021), yet many challenges remain in adopting generative AI models
such as ChatGPT within educational settings more generally.
While these advancements may seem to be ‘breakthroughs’, generative AI is still in a nascent stage.
Similar to any breakthrough technology, the introduction of AI models embodies many biases, data
accuracy problems, cognitive hallucinations and failures. While the progression of AI has the potential to
revolutionize many diverse fields of ontology such as in educational settings and in research domains more
generally, managing the quality of outputs remains a work in progress given how generative models can
create chatbots, deep fakes, movie dubbing, scripting emails/formal content, creation of art/videos and
others.
In summary, the introduction of ChatGPT grew remarkedly at the juncture of the receding pandemic with
its innovative features advanced quickly into many industry and related fields such as in education settings.
However, as the technology has evolved, many new challenges and ethical anomalies can now be
identified bringing to the forefront emerging paradoxes that are not easily solved. Our discussions thus far
suggest that generative AI tools are disruptive yet represent a transformative technological intervention.
However, generative AI tools such as ChatGPT are currently being debated and scientifically explored as
technology experts and advanced users consider ways that it can be universally adopted. In what follows,
the authors delve into the key challenges that threaten its adoption with the discussions focused mainly on
the education sector.
Related Local Literature
CONCEPTUAL FRAMEWORK

For a clear understanding of the concept of this study, the conceptual framework is presented in Figure 1.

This study aims to determine the the impact of artificial intelligence (AI) on the academic performance of

the criminology students at Palawan State University Main Campus.

 INPUT PROCESS OUTPUT

The varying extent of how

criminology students at
Impact of Artificial
Palawan State University Proposed plan of action
Intelligence to the
can use Artificial
Figure 1. The Conceptual Framework of the Study.
to minimize the potential
Academic Performance
Intelligence (Al) tools and negative impacts of
of criminology students
resources for their artificial intelligence on
at Palawan State
academic work in terms academic performance
University
of: and ensure it remains
• Determining the beneficial tool for
1. Frequency
correlation between the criminology students at
2. Purpose students' utilization of Al Palawan State University.
academic tools and their
• The level of academic
academic performance.
performance among
criminology students at
Palawan State University
who have access and
utilize artificial
intelligence in terms of:
1. Writing Skills
2. Critical Thinking Skills
Research Paradigm:

INDEPENDENT VARIABLE DEPENDENT VARIABLE

The underlying factors that can affect

Academic Performance of Criminology
students in using Artificial
Students
Intelligence AI

Figure 1. Paradigm of the Study

Research Hypothesis

Ho: There is no significant relationship between the factors that can affect students who use Artificial

Intelligence.

Ha: there is significant relationship between the factors that can affect students who use Artificial

Intelligence.
 CHAPTER III

RESEARCH METHODOLOGY

In this chapter, the researcher outlines the steps of the method used in the study. This includes;
Research Design, Research Locale, Population and Sample, Research Instrumentation, Data Collection
Procedure, Statistical Treatment.

RESEARCH DESIGN

The research design chosen for this study was descriptive survey design, which aims to
systematically collect data and describe the characteristics, features, or facts about a given population
(Nworgu, 2015). This design was selected to gather data on the perceived effect of artificial intelligence
(AI) tools on the academic performance of students in College of Criminal Justice Education at Palawan
State University. The data collected from the study population will be utilized to make informed
generalizations.

RESEARCH LOCALE

This study will be conducted in Palawan State University, the first state university in Palawan and
in Region IV. This school comprises Eight (8) Colleges which are the College of Criminal Justice
Education, College of Business and Accountancy, College of Hospitality and Tourism Management,
College of Arts and Humanities, College of Engineering, Architecture and Technology, College of Nursing
and Health Sciences, College of Sciences, and College of Teacher Education. It is located at Tiniguiban
Heights, Puerto Princesa City, Palawan (Latitude: 9.7773° N, Longitude: 118.7338° E).
POPULATION AND SAMPLE

he target population of
the study involves
Grade 12 STEM
(Science,
Technology, Engineering
and Mathematics) Senior
High School Students in
total
of 107 students
specifically STEM 1 with
35, STEM 2 with 32
students and STEM
3 with 40 students. Using
Slovin’s formula the
researchers calculated
the sample
size from 107. The
researchers got 84 as the
sample size. Hence, in
answering
the questionnaire at least
84 students answered
questionnaires are
expected to
return.
Among the specific
random/ probability
sampling techniques, the
study
used simple random
sampling to determine
the participants. This
sampling wa
he target population of
the study involves
Grade 12 STEM
(Science,
Technology, Engineering
and Mathematics) Senior
High School Students in
total
of 107 students
specifically STEM 1 with
35, STEM 2 with 32
students and STEM
3 with 40 students. Using
Slovin’s formula the
researchers calculated
the sample
size from 107. The
researchers got 84 as the
sample size. Hence, in
answering
the questionnaire at least
84 students answered
questionnaires are
expected to
return.
Among the specific
random/ probability
sampling techniques, the
study
used simple random
sampling to determine
the participants. This
sampling wa
he target population of
the study involves
Grade 12 STEM
(Science,
Technology, Engineering
and Mathematics) Senior
High School Students in
total
of 107 students
specifically STEM 1 with
35, STEM 2 with 32
students and STEM
3 with 40 students. Using
Slovin’s formula the
researchers calculated
the sample
size from 107. The
researchers got 84 as the
sample size. Hence, in
answering
the questionnaire at least
84 students answered
questionnaires are
expected to
return.
Among the specific
random/ probability
sampling techniques, the
study
used simple random
sampling to determine
the participants. This
sampling wa
he target population of
the study involves
Grade 12 STEM
(Science,
Technology, Engineering
and Mathematics) Senior
High School Students in
total
of 107 students
specifically STEM 1 with
35, STEM 2 with 32
students and STEM
3 with 40 students. Using
Slovin’s formula the
researchers calculated
the sample
size from 107. The
researchers got 84 as the
sample size. Hence, in
answering
the questionnaire at least
84 students answered
questionnaires are
expected to
return.
Among the specific
random/ probability
sampling techniques, the
study
used simple random
sampling to determine
the participants. This
sampling wa
The target population of the study involves all levels of Criminology Students in total of 254
respondents. Among the specific random sampling techniques, the study used convenience random
sampling to determine the participants. This sampling was used to ensure a fairly equal selection on the
population size and in accordance to the requirements of the research study.

RESEARCH INSTRUMENT

Construction. A questionnaire, will be constructed to ensure clarity, relevance, and effectiveness in

eliciting responses from the respondent. Careful consideration will be given to the wordings, avoiding
ambiguity or bias. This initial phone focuses ensuring the content of the questionnaire is relevant, clear,
and effectively addresses the research questions at hand.

Content Validation. To validate the content of the questionnaire, the researcher will submit it to
their research teacher for feedback, incorporating any necessary corrections or improvements.
Furthermore, validation will be sought from another experienced research leacher within the school, that
the questions effectively measure the variables of interest and align with the study’s objectives.

Grammar Checked. The finalized questionnaire will undergo a thorough grammar check to
eliminate any grammatical errors, ensuring that the language used is precise, clear, and grammatically
sound

DATA COLLECTION PROCEDURE

The researchers will seek permission from the school president and obtaining consent from
student. Once approvals are secured, the researchers craft a comprehensive questionnaire tailored to the
study’s objectives, ensuring relevance and clarity. This questionnaire undergoes a feedback process, with
the research teacher and validation from another experienced research teacher within the school. With a
validated questionnaire the researchers proceed to survey 100 randomly selected respondents. The
researchers assured the respondents the confidentially of the information that they will provide. The
questionnaires were collected right after they had been answered by the respondents. The gathered date
was tallied, analyzed and interpreted.

STATISTICAL TREATMENT
 To interpret the date effectively, the researchers will employ the following statistical treatment.
The Percentage, Weighted Mean, and Pearson correlation coefficient are the tools use to interpret data.

Percentage. This will employ to determine the frequency counts and percentage distribution of
personal information of the respondents.

Formula:

P=f/N x 100

Where:

P = percentage

f = frequency

n = total number of respondents

100 constant value

Weighted Mean. This will be employed to determine the students utilization extent of Artificial
Intelligence tool and level of academic performance
CONCEPTUAL FRAMEWORK
For a clear understanding of the concept of this study, the conceptual framework is presented
in Figure 1. This study aims to determine the the impact of artificial intelligence (AI) on the academic
performance of the criminology students at Palawan State University Main Campus.

INPUT PROCESS OUTPUT

The varying extent of how

criminology students at
Impact of Artificial
Palawan State University Proposed plan of action
Intelligence to the
can use Artificial
Figure 1. The Conceptual Framework of the Study.
to minimize the potential
Academic Performance
Intelligence (Al) tools and negative impacts of
of criminology students
resources for their artificial intelligence on
at Palawan State
academic work in terms academic performance
University
of: and ensure it remains
• Determining the beneficial tool for
1. Frequency
correlation between the criminology students at
2. Purpose students' utilization of Al Palawan State University.
academic tools and their
• The level of academic
academic performance.
performance among
criminology students at
Palawan State University
who have access and
utilize artificial
intelligence in terms of:
1. Writing Skills
2. Critical Thinking Skills
Research Paradigm:

INDEPENDENT VARIABLE DEPENDENT VARIABLE

The underlying factors that can affect

Academic Performance of Criminology
students in using Artificial
Students
Intelligence AI

Figure 1. Paradigm of the Study

Chapter 3 links
https://www.studocu.com/ph/document/central-luzon-state-university/introduction-to-entrepreneurship/
chapter-iii-methodology-research-locale/40217010
https://www.ijiras.com/2024/Vol_11-Issue_2/paper_3.pdf
Using Simple Random Sampling the researchers calculated the sample size from 100. The researchers got
84 as the sample size. Hence, in answering the questionnaire at least 84 students answered questionnaires
are expected tore turn.