ChatGPT integration significantly boosts personalized learning outcomes: A Philippine study

This study investigated the impact of AI integration, specifically ChatGPT, on personalized learning involving 785 college students in the Philippines who took the online survey. Utilizing regression analysis and an Omnibus ANOVA test, the study examined the influence of AI Integration alongside demographic variables such as age, sex, educational level, and type of school on personalized learning. Results indicate that AI integration can explain a substantial portion of the variability in personalized learning outcomes (approximately 88.54%). Specifically, ChatGPT demonstrates a significant positive effect on personalized learning, suggesting that as ChatGPT integration increases, personalized learning experiences also increase. However, demographic variables such as age, sex, educational level, and type of school show minimal effects on personalized learning outcomes, except for a potential trend for higher scores in private universities and colleges compared to state universities and colleges. These findings underscore the pivotal role of AI technologies, like ChatGPT, in enhancing personalized learning experiences while highlighting the need for further exploration of contextual factors influencing educational outcomes. The implications extend beyond the study to offer insights for educational stakeholders and policymakers, emphasizing the potential benefits of AI-driven personalized learning initiatives. However, limitations such as sample characteristics, measurement bias, and technology accessibility should be addressed in future research endeavors to maximize the benefits of AI integration in education.

Personalized learning through AI integration has emerged as a promising approach to tailor education to the individual needs of learners (Rouhiainen, 2019;Katiyar et al., 2024;Bhutoria, 2022;Kamalov et al., 2023;Gligorea et al., 2023).Existing literature has highlighted the potential benefits of AI-driven personalized learning, emphasizing increased learner engagement (Sabzalieva & Valentini, 2023), enhanced retention rates, and improved academic performance (Jian, 2023;Ayeni et al., 2024).However, despite these advancements, there remains a critical gap in understanding the precise methodologies and mechanisms necessary to optimize the integration of AI into personalized learning frameworks (Asirit & Hua, 2023).
This study aimed to gather empirical evidence on the impact of AI integration, specifically the use of ChatGPT, on learners' personalized learning experiences.The research sought to contribute basis to the development of effective teaching strategies and interventions aimed at improving students' educational outcomes.By examining the relationship between AI integration and indicators of personalized learning, the study aimed to determine if a significant linear relationship exists between the two variables.Additionally, the regression model employed in the study involved regressing the dependent variables representing indicators of personalized learning on the independent variables representing AI integration, while controlling for demographic variables.This approach allowed for an examination of the strength and direction of the relationships between the variables while considering the potential influence of demographic factors.

ChatGPT in Learning
The integration of ChatGPT in educational settings has sparked significant interest due to its potential to revolutionize learning experiences.Fütterer et al. (2023) delve into global reactions to ChatGPT's release, uncovering extensive discussions among educators regarding its advantages and concerns.Understanding these reactions is pivotal for identifying opportunities and challenges in integrating ChatGPT into education, underscoring the necessity for well-informed policy decisions and guidelines.For instance, Sabzalieva and Valentini (2023) demonstrate the diverse applications of ChatGPT in higher education, spanning teaching, learning experiences, research, and administration.While it enhances the learning process and streamlines administrative tasks, concerns about academic integrity, privacy, and accessibility highlight the importance of ethical regulation.However, ethically adapting ChatGPT can yield personalized learning experiences, administrative efficiency, research advancements, and community engagement, aligning with the objectives of the present study.Additionally, Margarella (2023) highlights ChatGPT's role as a virtual tutor, simplifying lesson planning and facilitating personalized interactions.Through structured prompts, educators can tailor interactions to individual needs, enriching dynamic learning environments, which resonates to explore personalized learning through AI integration in this study.
In Asirit and Hua's (2023) examination of AI awareness among college students in the Philippines, the findings underscored the importance of tailored AI education programs.These programs are seen as crucial in addressing knowledge gaps and preparing students for an AI-driven future.These insights directly inform the development of personalized AI education programs, aligning perfectly with the focus of the present study on personalized learning through AI integration.Furthermore, Rejeb et al. (2024) delve into public sentiment regarding ChatGPT's impact on education.They highlight benefits such as improved writing abilities and the creation of interactive learning environments.This supports the hypothesis that AI integration, particularly ChatGPT, can enhance personalized learning experiences.

Personalized Learning
In higher education, personalized learning has emerged as a critical strategy to address the diverse needs of students, enhance engagement, and improve learning outcomes.For example, Parikh (2023) argues that personalized learning empowers educators to tailor experiences, utilizing responsive learning management systems and asynchronous learning arrangements to accommodate various student demographics.Similarly, Hardy (2023) underscores the importance of personalized learning in meeting evolving student expectations and fostering engagement.The systematic reviews by Yuyun and Suherdi (2023) and Zhong (2022) delve into the key components and design elements of personalized learning, laying the groundwork for understanding its implementation.Furthermore, Claned (2024) explores the transformative potential of AI in personalized education, offering adaptive learning experiences and personalized instruction to deepen engagement and improve outcomes.
According to Ayeni et al. (2024), the integration of AI in education promises to revolutionize personalized learning.Through adaptive content delivery, intelligent tutoring systems, and other AI-driven technologies, personalized learning experiences are tailored to meet individual student needs, enhancing engagement and academic performance.The integration of AI in personalized learning presents new avenues for enhancing educational experiences.Gathering insights from empirical study, Dawes (2023) concludes that AI unlocks valuable insights into student behaviors and enhances teaching quality while Chawla (2024) suggested the transformative potential of generative AI by showcasing its ability to provide

Theoretical and Conceptual Framework
This study is grounded in the Cognitive Load Theory (CLT) by Sweller et al. (2011), which suggests that learning is influenced by the cognitive load imposed on learners.AI integration in learning such as the use of ChatGPT, can help manage the cognitive load by providing adaptive learning experiences tailored to individual student needs, thus being viewed to optimize learning efficiency (Tulsiani, 2024).Drawing from the tenets of CLT, this study examined intrinsic cognitive load through the complexity of learning content and the cognitive effort demanded for comprehension.Furthermore, the investigation into extraneous cognitive load delves into the efficacy of instructional strategies in alleviating cognitive burden, alongside an assessment of the clarity and coherence of instructional materials to minimize distractions.Moreover, explicit teaching strategies, such as explicit instruction and structured practice activities, are explored for their role in providing clear guidance and reducing cognitive load, particularly for novice learners (Sweller et al., 2011).These indicators shed light on the influence of AI integration on learners' cognitive load management and their overall learning experiences.
On the other hand, the Flow Theory as proposed by Csikszentmihalyi (1990), describes the state of deep engagement and immersion in an activity.Personalized learning experiences tailored to students' abilities and interests through AI integration can foster flow states, leading to enhanced learning outcomes and satisfaction (Rouhiainen, 2019).This theory encompasses the dependent variables, which include the measurement of flow experiences.This involves assessing participants' self-reported experiences of being in a state of flow during learning activities.Furthermore, optimal learning engagement is evaluated by assessing participants' levels of engagement, focus, and enjoyment during learning tasks.Finally, the quality of the learning experience is examined by exploring participants' perceptions of the overall effectiveness and quality of the learning process.
The research of Naik (2023) provides crucial insights into the transformative potential of personalized learning paths facilitated by AI, aligning with the core principles of the regression analysis framework.By grounding the study in a theoretical framework integrating personalized learning principles, cognitive psychology, and AI algorithms, Naik establishes a solid foundation for understanding the positive correlation identified between AI-driven personalized learning paths and improved academic performance, engagement, and retention underscores the relevance of AI integration in optimizing personalized learning experiences.

Research Design
This study employed a correlational research design (Sutradhar et al., 2023) to explore the relationship between AI integration, particularly ChatGPT, and personalized learning outcomes in higher education.Utilizing a cross-sectional approach (Wang & Cheng, 2020), data is collected at a single time point to assess how the variable are correlated.The primary variables in the linear regression analysis (Bevans, 2023) include ChatGPT utilization and personalized learning outcomes, measured by the flow of experience, learning engagement, and quality of learning.Demographic variables such as age, gender, educational level, and institution type are controlled to isolate the impact of AI integration on personalized learning outcomes.This is important because these factors can influence learning experiences, and controlling them ensures that the effects attributed to AI integration are not confounded by demographic differences (Hammer, 2011).

Respondents
The respondents for this study were college students in the Philippines through survey invitation email to participate (Lau, 2019).The selection was based on their fulfilment of the inclusion criteria; those who did not meet the criteria were excluded from data analysis Participants were informed as to the voluntary nature of the study.

Research Instrument
The research questionnaire utilized in this study consists of three parts: Part A focused on gathering respondents' demographic profiles, Part B assessed ChatGPT as the independent variable (IV) across all indicators including learning content, instructional design, delivery methods, and explicit instruction.This section utilizes a 5-point Likert scale ranging from "Poor" to "Excellent" for each indicator, with a total of 25 items (Mcleod, 2023).Part C evaluated personalized learning, comprising three components: A) flow experience, adopted from the Flow Short Scale (FSS) (Rheinberg et al., 2003), with 13 items rated on a scale of 1 to 7 indicating worry score from "not at all" to "very much"; B) learning engagement, and C) quality of learning, both of which are researcher-made measures.Each of these sections employs a 5-point Likert scale ranging from "very dissatisfied" to "very satisfied" for satisfaction assessment, with 12 items each (Mcleod, 2023).The validity of the instrument was ensured through expert validation involving five experts (Elangovan & Sundaravel, 2021), while reliability was established with an internal consistency alpha of .93 (Bobbitt, 2023).

Data Collection Procedure
Data collection involved administering an online survey through college research offices, with informed consent obtained from participants.Informed consent was documented via digital signatures using JotForm, ensuring that participants understood the study's purpose, procedures, and their rights.Respondents were screened based on inclusion criteria to ensure eligibility, specifically enrollment in SUCs, LUCs, or PUCs in the Philippines and prior use of ChatGPT as instructed by their professors in their learning tasks.Reminders were sent to nonresponders to boost participation rates.Eligible responses were included in the analysis based on the inclusion criteria.

Data Analysis
The data analysis procedure started with assumption checking to ensure the data meets linear regression requirements (Statisticslaerd, 2018).Upon confirmation of meeting assumptions, linear regression analysis is conducted to explore the relationship between AI integration aspects and personalized learning outcomes (Kanade, 2023).This analysis is performed using Jamovi statistical software (RCoreTeam, 2021;TheJamoviProject, 2022).
Finally, results are interpreted to understand the significance of AI integration on personalized learning outcomes in higher education.

Ethical Considerations
This study rigorously followed the ethical guidelines delineated by Williams (2023) to ensure the responsible and respectful gathering of data through surveys.In line with these principles, participants are provided with comprehensive information about the survey's

Results and Discussion
Integrating ChatGPT into educational settings holds the promise of revolutionizing personalized learning experiences (Carr, 2023).ChatGPT, an advanced AI model developed by OpenAI, offers educators innovative tools to tailor instruction and enhance individualized learning pathways.By simulating human-like conversation and providing intelligent responses, ChatGPT facilitates personalized support, immediate feedback, and expanded access to information.This section presents the results and discussion that delves into the empirical evidence and implications of ChatGPT on personalized learning, shedding light on its effectiveness and relevance in contemporary educational practices.
Table 1 presents the model fit measures for the regression analysis examining the impact of ChatGPT on personalized learning.The high R 2 value of 0.8854 suggests that a substantial portion, approximately 88.54%, of the variability in personalized learning outcomes can be explained by the integration of ChatGPT, along with other variables in the model.This underscores the pivotal role of AI integration in shaping personalized learning experiences, as highlighted by Zhai (2023) This implies that, within the context of the study, demographic variables such as age, sex, educational level, and type of school do not significantly influence the dependent variable.In other words, the personalized learning outcomes or the impact of ChatGPT on the dependent variable are not substantially affected by these demographic factors.Therefore, the effectiveness of personalized learning experiences facilitated by AI integration appears to be consistent across different demographic profiles, suggesting a degree of universality in its applicability.
and enhance engagement and quality.By leveraging AI, educators can offer adaptive learning pathways, personalized instruction, and instant feedback, ultimately revolutionizing traditional teaching methods and preparing students for an AIdriven future.

(
Dekkers et al., 2022).The study involved a sampling frame of 785 unduplicated students (262 State Universities and Colleges [SUCs], 260 Local Universities and Colleges [LUCs], and 263 Private Universities and Colleges [PUCs]), which exceeds the commonly accepted minimum sample size for linear regression analysis.According to de Longeaux (2021), a sample size of at least 500 is recommended to ensure robust and reliable regression estimates.The inclusion criteria encompassed students enrolled in any college courses in SUCs, LUCs, or PUCs, who have utilized ChatGPT in their learning tasks across online, hybrid, or asynchronous classes.
involvement, thereby obtaining informed consent.Upholding the paramount importance of confidentiality and anonymity, stringent measures, including the use of JotForm with encrypted data storage, are implemented to safeguard participants' privacy and ensure that individual responses remain secure and unidentifiable.Moreover, the survey design meticulously avoided bias and leading questions, maintaining neutrality to uphold the integrity of the data collected.Inclusivity across diverse demographics and backgrounds was also prioritized, ensuring a comprehensive representation of perspectives.Through transparent communication, participants were fully informed about the survey's purpose and the intended use of the data collected, fostering trust and credibility.

Table 1
Model Fit Measures , who emphasizes the transformative potential of AI in education.Moreover, the statistically significant F-statistic and low p-value (< 0.001) reinforce the reliability and validity of the regression model as a whole.This indicates that the model effectively captures the relationship between AI integration and personalized learning outcomes, aligning with Montenegro-Rueda et al. (2023), who assert that AI technologies, including ChatGPT, have a positive impact on teaching and learning processes.Additionally, the low root mean square error (RMSE) of 0.191 suggests that the regression model has good predictive accuracy.This means that educators can confidently utilize the model to inform decisions regarding the implementation of AI integration strategies aimed at improving personalized learning experiences for students.This echoes the findings of Albdrani and Al-Shargabi (2023), who demonstrate the potential of ChatGPT in providing personalized learning experiences, albeit with careful attention to ethical considerations.Overall, the findings underscore the significant potential of AI integration in enhancing personalized learning outcomes in educational settings.By leveraging AI technologies effectively, educators can create dynamic and tailored learning environments that cater to individual student needs, ultimately fostering improved student engagement, performance, and overall learning experiences.The Omnibus ANOVA test in table 2 was conducted to examine the collective impact of ChatGPT, age, sex, educational level, and type of school on personalized learning.The results indicate that ChatGPT significantly influences the dependent variable, as evidenced by a high F-value of 41.305 (p < .001).However, age, sex, educational level, and type of school do not demonstrate significant effects, with p-values above the commonly accepted threshold of .05.

Table 2
Omnibus ANOVA Test