The research presents the results of the evaluation of Vietnamese student’s perspectives on the effectiveness and challenges of using Artificial Intelligence for personalized learning ba
LITERATURE REVIEW
The concept of artificial intelligence (AI) originated in 1950, focusing on machines' ability to replicate and potentially exceed human intellectual capabilities Pioneers John McCarthy and Marvin Minsky posited that all aspects of learning and intelligence could be accurately represented, enabling computers to imitate these processes The Association for the Advancement of Artificial Intelligence further promotes this field, underscoring its significance in technological advancement.
Artificial Intelligence (AI) is defined as the scientific understanding of the processes behind thought and intelligent behavior, applied to computer systems (Sue Curry Jansen, 2022) AI systems can generate predictions, recommendations, or decisions based on specified human goals, influencing both real and virtual environments.
AI systems interact with us and influence our environment, both directly and indirectly They often operate autonomously and can adapt their behavior through contextual learning, as noted by UNICEF (2021).
Artificial Intelligence (AI) is now synonymous with advanced technology, leading to the development of concepts like 'machine learning' and 'deep learning' Unlike traditional views of AI as merely following set rules, modern AI utilizes deep learning algorithms to recognize patterns in data, often referred to as 'self-learning algorithms' (H Sheikh et al., n.d.).
AI is divided into three main categories based on machine complexity and intelligence: Artificial Narrow Intelligence (ANI), Artificial General Intelligence (AGI), and Artificial Super Intelligence (ASI) ANI, also known as weak AI, is designed for specific tasks and utilizes machine learning to gather extensive knowledge for precise execution It can mimic human actions within a limited scope, with notable examples including chess-playing programs like Deep Blue, language translation tools such as Google Translate, and advanced applications like self-driving cars and voice assistants like Siri and Alexa.
General Intelligence (AGI) is a form of strong AI that can perform any cognitive task a human can, including logical reasoning and a variety of general tasks, unlike Artificial Narrow Intelligence (ANI), which is limited to specific functions AGI operates autonomously, requiring no human intervention for tasks such as teaching, assigning classwork, and grading The next evolution, Artificial Super Intelligence (ASI), represents the pinnacle of AI development, featuring machine consciousness and the potential to surpass human intelligence across all domains ASI is expected to excel in complex problem-solving, creativity, and decision-making, exhibiting human-like emotions and senses, akin to AI robots.
Generative Artificial Intelligence (GenAI) has rapidly evolved, leading to the creation of various tools that enable users to generate images, videos, code, sounds, and more through simple keyword inputs Notable examples include ChatGPT, Midjourney, Dall-E 3, and deepfake technology GenAI aims to produce innovative and creative outcomes rather than just results based on existing data It shows exceptional inventiveness and quality in its solutions In higher education, GenAI is expected to serve as an intelligent tutoring system, personalizing learning materials to meet individual student needs by creating tailored instructional videos and visual aids This approach enhances students' understanding of complex concepts and generates assessments to evaluate their knowledge proficiency, ultimately improving their learning experience.
11 customized auditory feedback system can be developed using students' performance data to deliver precise instructions to each individual, hence enhancing student motivation and learning outcomes (Chan & Colloton, 2024)
Integrating intelligent tutoring system features, including an interactive learning environment and personalized feedback, significantly enhances personalized learning and deepens subject comprehension (Lâm & Vi, 2023).
AI applications in education enhance personalized learning through machine learning algorithms that analyze student data, including past performance and learning speed These algorithms detect patterns in behavior and achievements, allowing AI to tailor educational methods to individual needs By recommending appropriate materials and adjusting exercise difficulty, AI aims to provide targeted support, helping struggling students overcome challenges while encouraging high achievers to excel This customized approach fosters increased motivation and improved outcomes, as evidenced by Carnegie Learning's AI-driven math software, which has been shown to boost math proficiency by up to 30% (Harry, 2023).
Personalized learning emphasizes the importance of individual growth within a tailored educational environment, recognizing that learning is shaped by personal experiences, interests, and cultural backgrounds (Tapalova & Zhiyenbayeva, 2022) Each person's unique characteristics influence their approach to enhancing knowledge, critical thinking, and practical skills In contrast, traditional teaching methods impose a one-size-fits-all learning path, overlooking the diverse knowledge levels, motivations, and educational needs of students Therefore, evaluating the effectiveness of conventional educational methods requires consideration of the fundamental principles of human learning.
Implementing personalized learning approaches in education is crucial, as traditional methods do not cater to all students' needs In the classroom, some students thrive while others struggle academically, leading those who fall behind to seek alternatives like specialized classes or tutoring Tutoring represents a form of personalized learning, where students may choose between group settings or one-on-one sessions tailored to their specific needs This individualized attention, often found in smaller class sizes, allows tutors to adapt their teaching methods based on student performance However, the high costs associated with hiring tutors and accessing additional resources create disparities in educational opportunities, limiting access for those without financial means To address these challenges, integrating artificial intelligence into education could provide a more equitable solution, ensuring that all students receive the support they need to succeed academically.
Artificial intelligence, through big data analytics and machine learning, offers a cost-effective and convenient learning method tailored to students of different academic levels and sociometric backgrounds, ultimately enhancing their learning experiences and improving outcomes (Tapalova & Zhiyenbayeva, 2022).
An experiment was conducted by Tapalova and Zhiyenbayeva (2022) involving
A study involving 184 second-year students from Abay Kazakh National Pedagogical University and Kuban State Technological University aimed to explore the impact of AI technology on personalized learning in education Conducted over three months of remote learning from September to November 2021, and supervised by 20 IT academics, the research utilized various AI-based tools, including Altitude Learning, Gradescope, Knewton's Alta, Knowji, and Duolingo These tools facilitated independent study, evaluated student progress, and provided personalized feedback The findings revealed that most students found the AI tools valuable, emphasizing the importance of personalized learning pathways tailored to individual attributes and aspirations Students appreciated the immediate feedback and 24/7 access to resources, with 98% valuing the development of effective teaching strategies aligned with diverse contexts and knowledge levels.
Individuals highly value the ability to automatically monitor and measure their learning progress, alongside the comprehensive educational experiences that AI can provide This includes increased access to contemporary subjects and workplace skills, as well as the capability to analyze extensive training data quickly AI also supports remote social and emotional learning, offers immersive virtual learning environments, generates relevant online content, and aids in developing essential professional skills Research emphasizes the potential for educators to delve into the personalization of student learning through AI and highlights innovative educational activities that arise in the context of digital transformation in modern education.
Students' emotional well-being plays a crucial role in their academic success and serves as a key indicator of performance (Hưng, 2021) However, it is challenging for teachers to address the unique emotional needs of each student effectively Personalized learning approaches can provide timely and relevant instruction that aligns with individual emotional states By analyzing data on learning pace over time, educators can identify when students experience negative emotions, which can slow cognitive processing Artificial intelligence can then offer personalized recommendations based on each student's cognitive processing speed at any given moment.
RESEARCH FRAMEWORK
Research methodology
To investigate how Vietnamese students perceive the effectiveness and challenges of AI in personalized learning, a self-designed survey was conducted and distributed through digital platforms like electronic messages and Facebook This survey aimed to gather insights from students across various majors and years of study, including fields such as Economics, Language, Engineering, Information Technology, and more The survey consists of three main sections, focusing on students' experiences and perspectives regarding the use of AI-based learning applications during their university education.
21 questions that used to address Vietnamese students experiences and perspectives, specifically,
Section 1 - Information of students’ majors and years of study
In Section 2, students share their experiences with AI-based learning applications and platforms through a series of questions They are asked if they have utilized any AI learning tools, which specific platforms they are familiar with, and what features they have leveraged to enhance their learning The survey also explores whether these AI tools have encouraged a more proactive approach to their education Additionally, it assesses students' interest and readiness to engage with AI-based learning applications, as well as their beliefs regarding the efficacy of these technologies in improving their academic experiences.
AI ability to help them improve their learning efficiency and how they rate the
This article explores the effectiveness of AI-based learning platforms and applications by assessing students' actual experiences and usage The survey targets students who have utilized these AI tools, focusing on the features that enhance their learning and evaluating the AI's potential to facilitate proactive and efficient study habits By gathering insights from those with practical experience, the data collected will provide a more accurate representation of the impact of AI on the learning process, while filtering out responses from individuals who have not engaged with these technologies.
Section 3 explores students' perspectives on AI-based learning applications, focusing on their interest in personalized learning systems and comfort with AI integration in their education It highlights the significance of such systems in addressing individual needs and interests, as well as students' willingness to share learning outcomes and progress data Trust in AI's feedback, suggestions, and recommendations is also examined, alongside students' readiness to follow tailored learning paths designed by AI Additionally, the section assesses students' confidence in AI's capability to create effective learning experiences comparable to those of human educators Finally, it poses questions regarding desired features of AI-driven personalized learning systems and any concerns students may have about utilizing AI for personalized education.
The survey evaluates the effectiveness of AI-based learning platforms by examining key criteria, including students' interest and readiness to utilize these technologies, their belief in AI's capacity to enhance learning efficiency, and their enthusiasm for personalized learning systems designed to improve academic performance Additionally, it assesses students' comfort with universities employing AI to tailor the learning experience and the significance they place on personalized systems that cater to their individual needs.
The effectiveness of AI-based learning platforms in personalized education is influenced by various factors, including students' comfort with sharing learning data, trust in AI feedback, and willingness to follow AI-designed learning paths A positive correlation between these factors and the effectiveness of AI tools suggests that students perceive AI as beneficial for their learning experiences The author will utilize an analytical framework involving descriptive research, correlation analysis, and linear regression to assess the distribution of these criteria across different academic majors and years of study Various chart types, such as bar charts, histograms, scatter plots, and box plots, will be employed to visualize and interpret student perceptions Additionally, ANOVA analysis will test two hypotheses regarding the perceived effectiveness of AI in Vietnam's higher education, with a p-value threshold of 0.05 to determine the validity of the results This analysis aims to affirm students' perspectives on the effectiveness of AI-based personalized learning.
19 based learning platforms/applications The criteria “concerns orreservations about using
This article examines the challenges students encounter when utilizing AI-based learning platforms, focusing on their comfort level with sharing learning outcomes and progress data to assess their concerns about AI usage Additionally, it highlights the desired features of personalized AI learning systems that cater to students' unique needs, providing valuable insights into their preferences and informing policy recommendations.
The criteria established from the quantitative questions will serve as variables to evaluate their impact on the effectiveness of AI tools To analyze these criteria, the author will utilize a descriptive research method The accompanying visualization illustrates the relationships among the variables in the analytical framework.
Figure 2.1 Visualization of Analytical Frameworks
Explanation of variables in the Figure 2.1:
- S2.6 Effectiveness of AI-based learning tools (in learning process)
- S2.5 AI-based learning platforms/applications help students be more proactive in their learning process
- S2.8 AI-based learning platforms/applications’ features to improve students’ learning
- S3.10 AI-based personalized learning system’s expected features to suit student’s unique learning needs
- S3.11 Concerns or reservations about using AI in personalizing learning
- S2.3 Interest in using AI-based learning tools
- S2.4 Readiness to try AI-based learning tools
- S2.7 AI's ability to improve learning efficiency
- S3.3 Personalized learning system that suits individual needs
- S3.5 Trust in AI feedback, suggestions, and help
- S3.6 Trust in AI recommendations based on learning outcomes and progress
- S3.7 Willing to follow AI recommendations
- S3.8 Willing to follow AI-suggested learning path
- S3.9 AI's ability vs Human teachers in terms of designing effective learning experiences
- S3.1 Interest in using a personalized learning system
- S3.2 Acceptance of a personalized university learning system
- S3.4 Willing to share learning progress and outcomes data
Data collection process
The survey, aimed at university students, is distributed through various channels to gather responses from individuals in Vietnam who are either employed or pursuing a master's degree with an interest in the research topic To ensure data relevance, responses from 6 non-university students and 9 university students pursuing multiple majors will be excluded during the data cleaning process, resulting in a final sample size of 148 The collected responses are categorized by respondents' majors and academic years, enabling a thorough analysis of the data that highlights the unique characteristics and perspectives on AI-based learning applications across different fields of study and academic levels.
A survey of 148 students revealed that 95.3% have utilized AI-based learning applications or platforms Among the most popular tools, Chat GPT leads with 92.6% usage, followed by Quizlet at 64.9% and Duolingo at 62.8% Other notable applications include Grammarly (45.3%), Open AI (43.2%), and Coursera (16.9%), while Quillibot (13.5%) and Sololearn (2.03%) are less favored among Vietnamese students The survey included questions aimed at understanding their actual experiences with these AI learning tools.
A survey conducted among students revealed that AI significantly enhances proactive learning, with 23 first-year students affirming its benefits, alongside 41 second-year and 30 third-year students Only a small number of students expressed skepticism about AI's effectiveness—2 first-years, 4 second-years, 4 third-years, and 4 final-year students—while just 2 remained uncertain Various AI features have been identified as instrumental in improving students' learning experiences.
The article highlights key functionalities of educational tools, revealing that answering questions is the most utilized feature at 74.32% Additionally, suggesting ideas and summarizing curricula are significant at 58.78% and 54.73%, respectively Writing essays and proposals follows closely at 45.27% Other important features include providing solutions to math problems and correcting grammar errors, which account for 42% to 43% Evaluating student levels and learning progress is slightly lower at 37.16%, while mental health support features are also noted.
A recent survey indicates that 22% to 26% of students value features such as support and recommendations for improving learning and consolidating knowledge in AI-based learning applications Conversely, the least favored feature, recommending a unique learning path for students, received only 14.19% These findings highlight that the participating students, who have experience with AI learning tools, can offer valuable insights into the effectiveness and challenges of these platforms.
AI empowers educators to identify effective learning strategies for students By analyzing these insights, the author can gain a thorough understanding of Vietnamese students' viewpoints and provide valuable recommendations for practical policies that support the integration of AI applications in universities.
The study focuses on university students in Vietnam aged 18 to 22, representing Gen Z, a generation uniquely positioned in the digital age with a strong familiarity and comfort with new technologies This cohort demonstrates a proactive approach to adopting AI tools, surpassing other age groups in their engagement with technological advancements Survey findings confirm that these students are not only familiar with various AI-based learning tools but are also receptive to integrating them into their daily routines Additionally, they show a willingness to share data to optimize AI tools, which can be customized to meet their individual needs, ultimately providing direct benefits to their educational experiences.
RESEARCH FINDINGS
Descriptive analysis
Table 3.1 Descriptive Statistics for Main Variables
No Variables Mean Median Mode Standard Deviation
1 S2.6 Effectiveness of AI-based learning tools (in learning process)
2 S2.3 Interest in using AI-based learning tools 4.06 4 4 0.84
3 S2.4 Readiness to try AI-based learning tools 4.09 4 4 0.87
4 S2.7 AI's ability to improve learning efficiency 3.84 4 4 0.86
5 S3.3 Personalized learning system that suits individual needs 3.94 4 4 0.83
6 S3.5 Trust in AI feedback, suggestions, and help 3.89 4 5 0.98
S3.6 Trust in AI recommendations based on learning outcomes and progress 4.03 4 5 0.94
8 S3.7 Willing to follow AI recommendations 3.68 4 4 0.77
9 S3.8 Willing to follow AI- suggested learning path 4.01 4 5 0.99
10 S3.9 AI's ability vs Human teachers in terms of designing effective learning experiences
11 S3.1 Interest in using a personalized learning system 4.2 4 5 0.86
S3.2 Acceptance of a personalized university learning system 3.92 4 5 0.97
13 S3.4 Willing to share learning progress and outcomes data 3.88 4 4 0.9
Figure 3.6 Distribution of students at different majors
Analysis of the raw data reveals in figure 3.6 that despite the small sample size of
The distribution of scores varies significantly across different majors and years of study, with an even spread that incorporates diverse student perspectives However, majors like Transportation, Medicine and Pharmacy, Education, Biotechnology, Area Studies, Communication, and Social Sciences show limited responses, which may affect ranking results and hinder accurate forecasts Therefore, it is crucial to consider these factors in the data analysis section of the research.
Results of examining each variables and the colleration of each variables
a S2.6 Effectiveness of AI-based learning tools (in learning process)
Figure 3.7 Distribution effectiveness ratings for AI-based learning tools
Figure 3.7's histogram reveals that most ratings are concentrated at 4 or 5, with 4 being the most frequent by a significant margin This indicates a limited number of ratings at 1, 2, or 3, suggesting that few students view AI-based learning applications or platforms as ineffective or only moderately effective.
Figure 3.8 Distribution of ratings for the effectiveness for AI-based learning tools varies across different majors
According to figure 3.8, the highest mean effectiveness ratings for AI-based learning applications are observed in the majors of Transportation/Transportation Engineering (5.0), Construction - Architecture (4.33), and Language (4.01) Other notable fields with ratings above 4.0 include Biotechnology and Applications, Communication/Media/Digital Media, Education, Environmental and Natural Sciences, Health Care, Medicine and Pharmacy, and Social Sciences and Humanities, indicating that students in these areas find AI-based platforms particularly beneficial In contrast, majors such as Area Studies and Information Technology show lower average scores, suggesting that students in these disciplines may perceive AI-based platforms as less effective.
The findings highlight the academic majors that students perceive as most advantageous for AI-based learning systems Educational institutions and AI technology developers should consider prioritizing these majors and investigate the factors contributing to their higher effectiveness ratings relative to others.
Figure 3.9 Distribution of ratings for the effectiveness for AI-based learning tools by year of study
The histogram in figure 3.9 illustrates students' perceptions of AI-based learning tools, revealing a positive skew with most ratings falling between 3 and 5, indicating a favorable view of their effectiveness The most common rating is 4, signifying that a significant majority of students find these applications highly beneficial for their learning experience Only a small fraction rated them as 1 or 2, suggesting that few students consider AI-based tools ineffective Furthermore, the average effectiveness ratings across all academic levels range from approximately 3.7 to 4.2, demonstrating consistent appreciation for AI learning tools.
Final-year students rate AI-based learning applications/platforms with a mean effectiveness of around 4.2, suggesting they find these tools slightly more effective than their peers in earlier years In contrast, first-year and second-year students have lower average ratings of approximately 3.7 and 3.8, respectively, highlighting a trend where students in their final year perceive greater value in AI learning resources compared to those in their first and second years.
Although there are slight disparities in the average effectiveness evaluations between different years of study, these variances are largely insignificant These findings
33 indicate that students' opinions on the effectiveness of AI-based applications/platforms stay rather stable throughout their academic progression
Figure 3.10 Distribution of effectiveness for AI-based learning tools by majors and year of study
The effectiveness ratings for various majors show notable consistency across study years, as illustrated in figure 3.10 For instance, the "Economics" major has an average rating that increases from 3.62 for first-year students to 4.04 for final-year students However, some majors, like "Language," exhibit more significant fluctuations, with first-year students averaging 3.6 and third-year students reaching 5.0 Additionally, the sample sizes for each major-year combination vary widely, with some having only a few observations while others exceed 20 This variance in sample size is crucial to consider, as smaller samples may not accurately represent the broader population in the analysis of results.
Figure 3.11 Distribution of relationship between the number of observations and variability of effectiveness ratings
The analysis in Figure 3.11 reveals a clear trend: as the sample size increases, the variability in effectiveness ratings, indicated by the standard deviation, generally decreases This suggests that majors with a larger number of observations tend to show less variability in their judgments However, some majors with 20 to 30 observations still exhibit a standard deviation near 1.0, indicating exceptions to the trend The standard deviation for groups with multiple observations ranges from 0.0 to approximately 1.3 Additionally, the output table highlights the presence of several majors with only one observation, resulting in an undefined standard deviation, which are not represented in the plot due to their inability to be graphed.
The findings suggest that the variability in effectiveness ratings across different majors and academic years is affected by the number of observations, with larger groups typically showing reduced variability However, this relationship is not without exceptions, as some deviations from the expected pattern do occur Additionally, there is a growing interest in utilizing AI-based learning tools.
Figure 3.12 Distribution of interest ratings for AI-based learning tools
Students show significant interest in AI-based learning applications, with an average rating of 4.06 out of 5, as illustrated in figure 3.12 The most frequent rating is 4, supported by both the median and mode also being 4, indicating a strong overall enthusiasm for these platforms Although the standard deviation is 0.84, suggesting some variability in responses, most ratings are closely clustered around the mean, reinforcing the general trend of high student interest in AI learning tools.
The histogram reveals a skewed distribution towards the higher end of the rating scale, indicating a strong preference among students for AI-based learning applications and platforms The most common rating is 4, followed by 5 and 3, while ratings of 1 are notably less frequent.
The majority of students express a positive opinion towards AI-based learning applications, as evidenced by the clustering of ratings between 3 and 5, while only a small minority show minimal interest with ratings of 2.
In summary, according to this survey, students have a significant inclination towards utilizing AI-based learning applications/platforms, as the majority of them rate
36 them with scores of 4 or 5 out of 5 Nevertheless, there is variation in the degree of curiosity, which is probably influenced by individual preferences and experiences
Figure 3.13 Distribution of interest ratings for AI-based learning tools within each major
Figure 3.13 shows that most academic majors display a strong interest in AI-based learning applications, with ratings predominantly falling within the 3 to 5 range, reflecting a positive attitude towards these platforms.
Several majors, including Biotechnology, Communication/Media/Digital Media, and Construction - Architecture, demonstrate a high concentration of ratings at 4 and 5, indicating significant student interest However, the smaller sample sizes for these fields lead to less consistent rating distributions In contrast, majors like "Area Studies" and "Engineering" show a more balanced rating distribution, with a higher percentage of ratings in the 2-3 range, reflecting a broader range of interest levels Overall, despite variations in interest ratings across different majors, there is a clear trend of positive opinions towards AI-based learning tools in most fields.
Figure 3.14 Distribution of mean interest ratings for AI-based learning tools by years of study
The bar chart in Figure 3.14 depicts the average interest ratings for AI-based learning applications across different academic years Final-year students exhibit the highest mean rating, followed by second-year, third-year, and first-year students Notably, the median interest rating across all years is 4, indicating a consistently high enthusiasm among students for utilizing AI-based learning platforms.
Over the years, the dispersion of ratings has shown some fluctuation Final-year students have the highest average rating of 4.31 with a narrow range from 3 to 5, reflecting a consistently high level of interest In contrast, first-year students have the lowest average rating of 3.69 and a wider range from 2 to 5, indicating a greater diversity in interest levels Second-year students average 4.17, while third-year students average 3.88, both exhibiting similar rating ranges from 1 or 2 to 5 Notably, a few students in the second and third years reported exceptionally low-interest ratings, marking them as outliers.
In summary, this analysis indicates that students from all academic levels generally have a strong inclination toward utilizing educational resources based on artificial
Results of ANOVA analysis
Null Hypothesis (H0): Students in Vietnam's higher education do not perceive AI- based personalized learning as effective
Alternative Hypothesis (H1): Students in Vietnam's higher education perceive AI- based personalized learning as effective o S2.6
Since the p-value is below 0.05, we reject the null hypothesis (H0), indicating a statistically significant difference in how students in Vietnam's higher education perceive the effectiveness of AI-based personalized learning This means that students view AI-based personalized learning as an effective educational tool.
The p-value being less than 0.05 indicates a rejection of the null hypothesis (H0), demonstrating a statistically significant difference in the interest levels of students in Vietnam's higher education regarding AI-based learning platforms This implies that students exhibit diverse levels of interest in utilizing these innovative educational tools.
The p-value exceeding 0.05 indicates that we do not reject the null hypothesis (H0), implying no statistically significant difference in the readiness of students in Vietnam's higher education to adopt AI-based learning platforms Consequently, it can be concluded that these students demonstrate comparable levels of readiness towards utilizing AI-driven educational tools.
The p-value being less than 0.05 indicates that we reject the null hypothesis (H0), demonstrating a statistically significant difference in the perceptions of AI's role in enhancing learning efficiency among students in Vietnam's higher education This implies that students hold diverse beliefs regarding AI's potential to improve their academic performance.
Since the p-value exceeds 0.05, we do not reject the null hypothesis (H0), indicating that there is no statistically significant difference in the perceived importance of a personalized learning system that addresses individual needs and interests.
92 students in Vietnam's higher education In other words, students in Vietnam's higher education have similar views on the importance of a personalized learning system meeting their personal needs and interests o S3.5
The p-value being significantly below 0.05 leads us to reject the null hypothesis (H0), indicating a statistically significant difference in students' trust towards AI-generated feedback, suggestions, and assistance within Vietnam's higher education system This implies that students exhibit varying levels of trust in the support provided by AI.
With a p-value significantly below 0.05, we reject the null hypothesis (H0), indicating a statistically significant difference in trust towards AI recommendations among students in Vietnam's higher education, influenced by their learning outcomes and progress This highlights that students exhibit varying levels of trust in AI recommendations depending on their academic achievements and advancements.
With a p-value below 0.05, we reject the null hypothesis (H0), indicating a statistically significant difference in the willingness of students in Vietnam's higher education to follow AI recommendations and suggestions.
93 other words, students in Vietnam's higher education have varying levels of willingness to follow the recommendations and suggestions given by AI o S3.8
The significantly low p-value, below 0.05, indicates a rejection of the null hypothesis (H0), demonstrating a statistically significant difference in students' willingness to engage with AI-designed learning paths in Vietnam's higher education This finding highlights the diverse levels of commitment among students towards personalized learning experiences offered by AI platforms.
The p-value being significantly less than 0.05 leads us to reject the null hypothesis (H0), indicating a statistically significant difference in Vietnamese higher education students' beliefs regarding AI's ability to design effective learning experiences comparable to those created by teachers This highlights that students in Vietnam's higher education system possess diverse levels of confidence in AI's effectiveness in this area.
With a p-value below 0.05, we reject the null hypothesis (H0), indicating a statistically significant difference in the interest levels of Vietnamese higher education students towards a personalized learning system designed to enhance academic performance.
94 higher education have varying levels of interest in using a personalized learning system designed to help them achieve better academic results o S3.2
The p-value being less than 0.05 indicates a rejection of the null hypothesis (H0), signifying a statistically significant difference in comfort levels among students in Vietnam's higher education regarding the use of AI to personalize their learning experiences This implies that students exhibit diverse degrees of comfort with their universities implementing AI for personalized education.
A p-value of less than 0.05 indicates a statistically significant difference in the comfort levels of students in Vietnam's higher education when it comes to sharing data about their learning outcomes and progress This suggests that students exhibit varying degrees of willingness to share their information, which could enable AI to offer tailored suggestions and guidance aimed at enhancing their educational experiences.
The ANOVA analysis aimed to evaluate students' perceptions of AI-driven personalized learning in Vietnam's higher education system Key factors examined included students' willingness to adhere to AI recommendations, their commitment to AI-curated learning pathways, their confidence in AI's capability to create effective learning experiences, and their interest in utilizing AI-based learning platforms.
Students' attitudes toward AI-based learning systems vary significantly across different years of study, indicating that their experiences and exposure to AI technology may shape their perceptions and willingness to engage with these tools.