INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

The Influence of Performance Expectancy, Effort Expectancy, and

Social Influence on Artificial Intelligence Adoption Behaviour: A

Case Study from a Malaysian University

Pang Zong Wei¹, Karen Poon Yean Peng², Simranpreet Kaur Hansaram³, Aw Yoke Cheng⁴, Chong Kim

Loy^5*

^1,2University of Wollongong, Malaysia, 40150 Shah Alam, Malaysia

⁴Berjaya University College, 55100, Kuala Lumpur, Malaysia

^{3, 5}UNITAR International University, 47301 Petaling Jaya, Malaysia

*Corresponding Author

DOI: https://dx.doi.org/10.47772/IJRISS.2025.91100379

Received: 28 November 2025; Accepted: 04 December 2025; Published: 11 December 2025

ABSTRACT

This paper explores the variables that affect the use of Artificial Intelligence among students at the ABC

university at Malaysia, based on major constructs of Unified Theory of Acceptance and Use of Technology

(UTAUT). The study concentrates on three independent variables, which include the performance expectancy

(PE), effort expectancy (EE), and social influence (SE), on Artificial Intelligence (AI) adoption behavior. Data

was collected from a sample of 211 students at ABC University via an online questionnaire. The descriptive

statistics, reliability test, Spearman correlation were used to analyze the data. Results indicate high levels of

internal consistency of all the constructs and high positive associations between each variable and AI adoption

behavior. The strongest predictor was effort expectancy, which demonstrates the significance of AI systems

that are intuitive and easy to use. The social influence and performance expectancy were also found to play

significant roles. Meaning that students are both social-validation- and perceived-academic-benefit-motivated.

Keywords: Artificial Intelligence (AI) Adoption, Unified Theory of Acceptance and Use of Technology

(UTAUT), Performance Expectancy, Effort Expectancy, Social Influence

INTRODUCTION

Society has become more technology-driven, especially with the development of new technology. The process

of adopting new technology is based on two major factors the first one is the perceived ease of use and the

second, the perceived usefulness (Izham, H.I.B, et al., 2025). Among the newest technology that has become

popular is artificial intelligence (AI). It has made the lives of people convenient, but at the same time, has

gradually interfere with how industries work (Russell et al., 2021; Zhang et al., 2024).

For example, in the education sector, it has lead to a disruptive transformation (Păvăloaia & Necula, 2023).

With AI tools, students found real-time support, generate content, summarize reading text, address academic

challenges, and engage with data with natural language (Black & Tomlinson, 2025; Vieriu & Petrea, 2025).

This has upset the conventional education systems, especially in the tertiary education (Paek & Kim, 2021;

Ruano-Borbalan, 2025). Nevertheless, AI tools that are easy to use and readily accessible, are the factors

behind its increasing popularity among college students (Ho Ngoc Hai, 2023; Li et al., 2023).

The University of ABC is one of the institutions where students are rapidly incorporating AI into their

academic routines. The application of AI-based tools, like ChatGPT, is gaining popularity as students believe

that it enhances learning, knowledge, and writing development (Nazari et al., 2021; Nhu et al., 2024). Also,

ChatGPT can serve as a fast way to solve problems, meet deadlines and information processing. Moreover, AI

Page 4818

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

applications bring assistance 24/7 and provide opportunity to ask questions that students might feel

embarrassed to bring up in the classroom.

In this manner, AI comes in handy for introverted learners especially those whose first language is not English.

While these benefits explain why many students are turning to AI, their decision to adopt such tools is also

influenced by a variety of personal, social, and institutional factors (Granić, 2023; Strzelecki, 2023; Uzun,

2023; Yakubu et al., 2025). The ethical frameworks established by the technology developers (OpenAI, 2024)

and institutional concerns over academic integrity (Cornell University, 2023) further shape the adoption

environment.

Problem Statement and Research Gap

ABC is a university where a large population of local and international students are enrolled. Thus, the

dynamics of AI adoption might be different compared to those in Western universities (Faraon et al., 2025).

Nonetheless, little empirical data has been found regarding the perceptions of performance expectancy, effort

expectancy, and social influence in the adoption of AI among Malaysian students (He et al., 2024; Ruslan,

2024; Yakubu et al., 2025).

This paper aims to close this gap by analyzing the behavioral and contextual variables that can determine the

adoption of AI by ABC university students. Consequently, it also contributes to academic literature as well as

practical solutions to higher education institutions in Malaysia. Based on the preceding paragraphs, the

following are the research objectives and research questions of this study.

Research Objectives (RO)

RO1: To analyze the influence of performance expectancy on Artificial Intelligence adoption behavior among

University of ABC students.

RO2: To analyze the influence of effort expectancy on Artificial Intelligence adoption behavior among

University of ABC students.

RO3: To analyze the influence of social influence on Artificial Intelligence adoption behavior among

University of ABC students.

Significant of Study

The significance of this study is to offer customized insights that reflect the unique social, cultural, and

institutional realities of Malaysian universities. It can examine the roles of performance expectancy, effort

expectancy, and social influence in predicting AI adoption behavior among ABC university students,

LITERATURE REVIEW

This section draws upon well-known model frameworks, the Unified Theory of Acceptance and Use of

Technology (UTAUT) to analyze and provide evidence for the dependent and independent variables in the

research. To demonstrate the theoretical and practical significance of each component, this paper will review

previously published research findings. This leads to the formulation of research hypotheses and the suggested

study framework.

Identification and Conceptualization of Variables

AI Adoption Behavior (DV)

AI Adoption Behavior (AAB) refers to the degree to which college students use AI technology in their

academic or personal lives to provide them with information, generate content, or answer questions (Nhu et al.,

2024). As example, students use ChatGPT to write outline and add requirements and guidance for assignments

Page 4819

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

or projects. In doing so, they believe they get the most suitable answer (Sánchez-Prieto et al., 2020; Chong et

al., 2022; Wen et al.,2024).

Past studies suggest that students who use AI for writing and research are more satisfied and productive than

those who did not use AI (Nazari et al., 2021). The impact of AAB is also related to the usage environment. AI

is more suitable for highly repetitive work or time-consuming tasks such as complex data processing or large

volume translations (Holmes et al., 2021; Zhang et al., 2024).

Furthermore, AI has gained popularity among students because they leverage this powerful tool. However,

how much students utilize the tool depend on their feelings about its worth, usefulness, and dependability

(Cunningham, 1967; Chong et al., 2022; Bloomfield & Rushby, 2024). This link between perceived risk,

satisfaction, and behavioral intention is a recognized concept in literature (Tran, 2020).

Performance Expectancy (IV1)

Performance expectancy may be perceived as how useful technology can be in working or studying as

perceived by people (Merz et al., 2025) It is grounded in Unified Theory of Acceptance and Use of Technology

(UTAUT) and conceptually aligned with perceived usefulness in Technology Acceptance Model (TAM)

(Davis, 1989), and very much related to perceived utility (Hoo et al., 2023).

According to Mustafa and Garcia, (2021) the literature has indicated that performance expectancy is one of the

central variables in adopting and final usage of information systems where high-performance expectancy

creates more acceptability by a user in adopting the new technology (Jain et al., 2022). According to

Sewandono et al. (2022), performance expectancy is the evaluation of how individuals believe in technology

will be for their professions or educations.

Effort Expectancy (IV2)

Effort Expectancy (EE) refers to the ease or difficulty of using a particular technology (Du & Beibei Lv, 2024),

especially when first using any new technology (Du & Beibei Lv, 2024). Venkatesh et al. (2003) proposed this

concept in the Unified Theory of Acceptance and Use of Technology (UTAUT) and pointed out that one of the

four core determinants that affect users' willingness to use new technologies is Effort Expectancy. This concept

is similar to the perceived ease of use in the Technology Acceptance Model (TAM), where users prefer to

accept technologies that they think are easier to use and understand (Li et al., 2023).

Social Influence (IV3)

Social influence in this study refers to the phenomenon where an individual's thoughts, behaviors, and

decisions gradually conform to or follow the patterns of the general public in order to adapt to societal norms

(Spears, 2021). This theory is particularly applicable to the factors influencing students' adoption of AI in this

study. During their learning process, students are inevitably exposed to their teachers, classmates, and even the

educational system, all of which generally have their own rules.

According to Guassi Moreira et al., (2021), when these groups' thoughts and behaviors align, students' choices

can be influenced. Essiz & Mandrik (2021) also point out that family and friends are one of the most important

sources of information in the consumer decision-making process, which also proves that students' decision-

making judgments are influenced by the values of their family and friends.

To further illustrate social influence, this study found that Alessandro Rovetta et al. (2025) also indicate that

people tend to accept culture, values, and social identity of their social group and adjust their behavior

accordingly. This suggests that when people's decisions are disapproved by the public, their decisions will

gradually change to conform to the public's direction, which is similar to social identity theory (Tajfel &

Turner, (1979).

Page 4820

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Conceptual Framework

This study examines the relationship between AI Adoption Behavior (AAB) and three key factors derived from

the UTAUT model: Performance Expectancy (PE), Effort Expectancy (EE), and Social Influence (SI). Based

on the literature review and underpinning theory, the following is the theoretical framework for this study.

Figure 1 Conceptual Framework

Hypothesis Formulation

Relationship between Performance Expectancy and AI Adoption Behavior

Performance Expectancy (PE) can be described as a degree to which the students suppose that utilization of AI

can help them to improve their academic results (Moradi, 2025). Previous research has consistently shown that

PE and technology adoption have a positive correlation (Sewandono et al., 2022; Yakubu et al., 2025).

As an example, students are likely to use AI tools when they see apparent scholarly advantages like better

writing or more rapid access to information. Nonetheless, there is some research which indicates the strength

of such a relationship is not constant in different contexts; in the contexts where students do not trust AI-

generated information, PE might have a more modest impact on adoption intention (Li et al., 2023; Safdar et

al., 2024; Kiat et al., 2025;). Since the context of higher education in Malaysia is unique, the key research

question can be tested by whether PE has a significant impact on the usage of AI tools by students ABC

university.

H1: Performance Expectancy has a positive relationship with AI adoption behavior among ABC university

students.

Relationship between Effort Expectancy and AI Adoption Behavior.

Effort Expectancy (EE) is the ease of using AI tools in studying. Past research attests to the fact that students

tend to use technologies that are simple in nature (Venkatesh et al., 2003). Even in the example of ChatGPT, its

conversational interface enables even non-technical students to interact with AI, which also makes it adoptable

(Strzelecki, 2023).

However, certain studies show that in the case of younger and more technologically savvy groups, the impact

of EE is not as strong, since students assume that most tools should be easy to use (Li et al., 2023). This is

where an issue emerges whether EE plays significant role in adoption in the university setting like ABC

university which have students that are already familiar with digital platforms.

Page 4821

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

H2: Expectancy has a positive relationship with AI adoption behavior among ABC university students.

Relationship between Social Influence and AI Adoption Behavior.

Social influence refers to the influence of friends, family, classmates, and other important people who

frequently interact with each other about use of AI technology. When this happen, it will also enhance a

person's view on the use of this technology (Cheng et al., 2022; Jain et al., 2022). Studies have shown that

people tend to follow the technology choices of those around them to meet social expectations, so social

expectations will greatly affect individual technology use behavior (Ruslan, 2024; Merz et al., 2025).

For example, if someone is around a person who is willing to accept AI technology, then there is a likelihood

they can influence each other to use the technology. On the contrary, if culture of a certain art college is to

support original design, then they are likely to be influenced by the college not using AI technology as their

own creative content (Kraatz & Xie, 2023).

H3: Social influence has a significant positive relationship with AI adoption behavior among ABC university

students.

METHODOLOGY

This study employed a quantitative research approach (Hunziker & Blankenagel, 2024) grounded in the

Unified Theory of Acceptance and Use of Technology (UTAUT) to investigate the factors influencing AI

adoption behavior among students. A sample of 211 undergraduate and postgraduate students from a single

university was obtained via an online questionnaire using a non-probability convenience sampling method

(Andrade, 2021; Rahman, 2023). The data collection ensured ethical standards, including informed consent

and voluntary participation (Millum & Bromwich, 2021; Mumford, Higgs, & Gujar, 2021; Hansaram &

Munap, 2025).

The instrument measured the independent variable which are Performance Expectancy, Effort Expectancy, and

Social Influence and the dependent variable, AI Adoption Behavior. The survey used a five-point Likert scale

(Mohd Rokeman, 2024). Data were analyzed using SPSS version 29 (Pallant, 2020), employing descriptive

statistics, reliability analysis (Cronbach, 1951; Hair et al., 2019), and Spearman correlation (Shapiro & Wilk,

1965) to examine relationships between the constructs at a significance level of p < 0.05.

RESULTS AND DISCUSSIONS

This section presents the analysis of the survey data collected to investigate the factors influencing AI adoption

behavior among students. The findings are structured as follows: demographic profile of respondents,

assessment of measurement reliability, descriptive statistics of the constructs, correlation analysis, and

hypothesis testing.

Demographic Profile of Respondents

Table 4.1 Demographic profile

Demographic Characteristic

Gender

Category

Female

Frequency(n) Percentage (%)

125

86

59.20%

40.80%

100.00%

64.90%

16.10%

Male

Total

211

137

34

20–22 years old

23–25 years old

Age

Page 4822

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Above 25 years old

Below 20 years old

Total

12

5.70%

13.30%

100.00%

47.4

28

211

100

69

Bachelor’s Degree

Diploma

Programme of study

32.7

Foundation

22

10.4

Postgraduate

20

9.5

Total

211

44

100.00%

20.9

Academic writing / assignments

Entertainment / casual use

Idea generation / brainstorming

Learning / understanding difficult topics

Translation or language practice

Total

Main

purpose

of

Using

ChatGPT

14

6.6

74

35.1

37

17.5

42

19.9

211

100

From table 4.1, the online survey yielded 237 responses, of which 211 were valid and complete, resulting in a

high response rate of 89.11%. This sample size of 211 exceeds the recommended minimum ratio of 10:1

(respondents to items) for statistical reliability, given the 18 measurement items used in the study (Memon et

al., 2020; Hair et al., 2021). The demographic analysis reveals a diverse respondent profile. The sample

comprised slightly more females (59.2%) than males (40.8%). In terms of age, most respondents (64.9%) were

between 20 and 22 years old. Academically, bachelor’s degree students formed the largest group (47.4%),

followed by Diploma (32.7%), Foundation (10.4%), and Postgraduate (9.5%) students.

The field of study was well-distributed, with the largest cohorts from Communication/Media (30.8%) and

Business/Management/Marketing (26.5%). Regarding the primary purpose of using ChatGPT, the most

common reason was idea generation and brainstorming (35.1%), followed by academic writing/assignments

(20.9%) and translation/language practice (19.9%), indicating a predominant use for educational and

productive tasks (Black & Tomlinson, 2025).

4.2 Reliability Analysis

Table 4.2 Reliability Analysis (Cronbach's Alpha)

Variable

Cronbach’s Alpha Value

No. of items

Performance Expectancy

Effort Expectancy

Social Influence

0.848

0.869

0.857

0.784

6

AI Adoption Behavior

Page 4823

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

The internal consistency of the measurement scales was assessed using Cronbach's Alpha (Cronbach, 1951).

As shown in Table 4.2, all constructs demonstrated high reliability, with values exceeding the accepted

threshold of 0.7 (Hair et al., 2019). The independent variables: Performance Expectancy (α = 0.848), Effort

Expectancy (α = 0.869), and Social Influence (α = 0.857) and the dependent variable, AI Adoption Behavior (α

= 0.784), all exhibited good to excellent reliability, confirming the instrument's internal consistency.

Descriptive Statistics of Constructs

Table 4.3 Descriptive Statistics

Code

Item

Sample

size(n)

Mean

Standard

Deviation

Performance Expectancy (PE)

PE1

Using ChatGPT in my job/assignment would enable me to 211

4.2

1.038

accomplish tasks more quickly.

PE2

PE3

PE4

PE5

PE6

Using ChatGPT would improve my academic performance.

Using ChatGPT for my job would increase my productivity.

Using ChatGPT would enhance my effectiveness on my job.

Using ChatGPT would make it easier to do my job.

I would find ChatGPT useful in my job.

211

3.83

4.15

3.93

4.06

4.03

1.125

1.043

0.993

1.074

1.028

Effort Expectancy

EE1

EE2

EE3

EE4

EE5

EE6

Learning to operate ChatGPT would be easy for me.

211

4.18

3.93

3.92

4.01

3.96

3.87

1.034

1.033

1.174

1.067

0.999

1.134

I would find it easy to get ChatGPT to do what I want it to do.

My interaction with ChatGPT would be clear and understandable.

I would find ChatGPT to be flexible to interact with.

It would be easy for me to become skilful by using ChatGPT.

I would find ChatGPT easy to use.

Social Influence

SI1

SI2

SI3

SI4

SI5

People I care about encourage me to use ChatGPT.

211

3.97

3.88

3.73

3.8

1.062

1.097

1.261

1.064

1.083

Most people surrounding me use ChatGPT.

My lecturers’ opinions influence my decision to use ChatGPT.

People who influence me encourage me to use ChatGPT.

The university environment creates pressure for me to adopt 211

ChatGPT in learning.

3.91

SI6

Using ChatGPT is viewed positively by people whose opinions I 211

value.

3.79

1.17

AI Adoption Behavior

Page 4824

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

AAB1

AAB2

AAB3

AAB4

AAB5

AAB6

I will continue to acquire ChatGPT related information.

I will keep myself updated with the latest ChatGPT applications.

I intend to use ChatGPT to assist with my learning.

I will continue to learn ChatGPT.

211

4.07

3.89

4.14

4.11

4.12

4.45

1.117

1.153

0.959

1.025

0.983

0.817

I frequently use ChatGPT to complete academic-related tasks.

I plan to integrate ChatGPT into my long-term study routine.

The descriptive statistics for the key constructs revealed consistently positive perceptions among respondents.

For Performance Expectancy (PE), the mean scores for the individual items ranged from 3.83 to 4.20 on a 5-

point scale. The strongest agreement was with the statement, "Using ChatGPT would enable me to accomplish

tasks more quickly" (M=4.20, SD=1.038), underscoring the tool's perceived role in enhancing efficiency

(Sewandono et al., 2022).

Similarly, for Effort Expectancy (EE), mean scores fell between 3.87 and 4.18, with the highest score

attributed to "Learning to operate ChatGPT would be easy for me" (M=4.18, SD=1.034), indicating a

consensus on the platform's learnability (Li et al., 2023). The Social Influence (SI) construct, while displaying

slightly lower mean scores from 3.73 to 3.97, still reflected a positive social environment, with the strongest

influence coming from close peers as captured by the item, "People I care about encourage me to use

ChatGPT" (M=3.97, SD=1.062) (Spears, 2021).

Finally, the dependent variable, AI Adoption Behavior (AAB), demonstrated the most robust results, with all

items exceeding 3.89. The highest level of agreement was found for the intention, "I plan to integrate ChatGPT

into my long-term study routine" (M=4.45, SD=0.817), signaling a strong commitment to the tool's continued

use (Holzmann et al., 2025).

Normality Test Result

Table 4.4 Normality Test

Variables

Kolmogorov-Smirnov^a

Shapiro-Wilk

Statistic

0.230

df

Sig.

Statistic

0.875

df

Sig.

211

<.001

211

<.001

Performance Expectancy

Effort Expectancy

0.293

0.849

0.285

0.838

Social Influence

0.213

0.864

AI Adoption Behavior

a. Lilliefors Significance Correction

To assess the distribution of the data, normality tests were conducted. The Shapiro-Wilk test indicated a

significant deviation from normality for all constructs (p < .001), as shown in Table 4. Consequently, non-

parametric statistical methods, specifically Spearman's rank-order correlation, were employed for the

subsequent inferential analysis to ensure robust and valid findings (Shapiro & Wilk, 1965; Mat Roni &

Djajadikerta, 2021).

Page 4825

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Correlation Analysis

Table 4.5 Correlation analysis

Variables

Performance

Expectancy

1

Effort

Expectancy

.797^**

<.001

211

Social

AI Adoption

Influence Behavior

Spearman's Performance Correlation Coefficient

rho

.761^**

<.001

211

.750^**

<.001

211

Expectancy

Sig. (2-tailed)

.

N

211

Effort

Correlation Coefficient

.797^**

<.001

211

1

.757^**

<.001

211

.770^**

<.001

211

Expectancy

Sig. (2-tailed)

.

N

211

Social

Correlation Coefficient

.761^**

<.001

211

.757^**

<.001

211

1

.725^**

<.001

211

Influence

Sig. (2-tailed)

N

.

211

AI Adoption Correlation Coefficient

.750^**

<.001

211

.770^**

<.001

211

.725^**

<.001

211

1

Behavior

Sig. (2-tailed)

.

N

211

**. Correlation is significant at the 0.01 level (2-tailed).

Spearman’s rank-order correlation was used to examine the relationships between the independent variables

(PE, EE, SI) and the dependent variable (AAB). The results, summarized in Table 4.5, reveal significant

positive correlations between all three independent variables and AI Adoption Behavior (Jain et al., 2022).

Specifically, AI Adoption Behavior has a strong positive correlation with Effort Expectancy (ρ = 0.770, p <

.001), followed by Performance Expectancy (ρ = 0.750, p < .001) and Social Influence (ρ = 0.725, p < .001).

Furthermore, the independent variables were also highly correlated with each other, with the correlation

between PE and EE (ρ = 0.797) indicating potential multicollinearity, which should be diagnosed in future

regression analyses.

Hypothesis Testing

Table 4.6 Hypothesis Testing Results

Hypothesis Description

Correlation

p-value

Result

Coefficient (ρ)

There is a significant positive relationship between 0.75

< 0.001

H1

Accepted

Performance

Behavior.

Expectancy

and

AI

Adoption

Page 4826

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

There is a significant positive relationship between 0.77

Effort Expectancy and AI Adoption Behavior.

< 0.001

H2

H3

Accepted

There is a significant positive relationship between 0.725

Social Influence and AI Adoption Behavior.

The study proposed three hypotheses to test the positive relationship between the independent variables and AI

Adoption Behavior. The results of the Spearman correlation analysis, presented in Table 4.6, led to the

acceptance of all three hypotheses. The findings confirm that students' adoption of AI tools like ChatGPT is

significantly influenced by their perception of its performance benefits (PE), ease of use (EE), and the social

pressures and norms surrounding its use (SI) (Jain et al., 2022).

The analysis provides robust evidence that Performance Expectancy, Effort Expectancy, and Social Influence

are significant determinants of AI Adoption Behavior among university students. The measurement instrument

proved reliable, and all hypothesized relationships were strongly supported. The high inter-correlations among

the independent variables suggest a intertwined perception of the tool's usefulness, ease of use, and social

acceptance. These findings offer valuable insights for educators and institutions seeking to foster the

productive adoption of AI in educational contexts.

CONCLUSION

This section focuses on discussion that reveals the study's research findings. The primary objective is to

hypothesize the results and assess them in light of the study objectives and questions. Specifically, is to

examine how Performance Expectancy (PE), Effort Expectancy (EE), and Social Influence (SI) influence AI

Adoption Behavior (AAB) among students at ABC University, Malaysia.

Beyond interpreting the key results, the study also highlights the theoretical and practical implications of the

findings, discusses the limitations encountered in the research process, and offers recommendations for future

studies. By doing so, it provide a clearer understanding of the factors shaping students’ intention to use

ChatGPT, as well as guidance for educators, institutions, and researchers who intend to explore AI adoption in

higher education.

DISCUSSION OF FINDINGS

This section is an inquiry into how the research findings are interpreted, how the statistical results are

connected to the research questions and the underlying UTAUT framework (Venkatesh et al., 2003). This

discussion will provide the reasons as to why the identified relationships between Performance Expectancy,

Effort Expectancy, Social Influence, and AI Adoption Behavior are present, by putting them into perspective in

existing literature.

Relationship between Performance Expectancy and AI Adoption Behavior

The results of the analysis affirm that there is strong positive correlation between performance expectancy and

the behavior of AI adoption among students in ABC university. This result means that the students who hold

the opinion that the application of AI tools such as ChatGPT will render them positive academic benefits are

much more prone to the integration of this technology into their learning activities. Close relationship

demonstrates the power of perceived utility as one of the main factors of technology acceptance (Davis, 1989).

This finding is closely related to the central principles of the UTAUT model and is backed by the current

studies in the field of educational AI. An example of this is that in a study by Yakubu et al. (2025) on the use of

generative AI to learn, the performance expectancy was the best predictor of behavioral intention. Their

investigation concluded that students are encouraged to embrace AI when they believe that it provides them

with a direct tool to improve their performance and output in scholarly institutions, which is the direct finding

that aligns with the findings of this paper.

Page 4827

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Further support comes from Moradi (2025), who investigated ChatGPT acceptance among Chinese university

students. In the study, the perceived usefulness of AI on enhancing learning outcomes came out as a crucial

issue that affected adoption, which agreed with the findings of the ABC university students. Thus, the findings

of this paper, in line with the literature, prove that the instrumental value of AI in promoting academic efficacy

is a crucial factor that predetermines its application in higher education.

Relationship between Effort Expectancy and AI Adoption Behavior

The research develops a positive strong correlation between the expectancy of effort and the adoption behavior

of AI. It means that the convenience of using AI tools is one of the key facilitators of their adoption. As soon as

students discover AI applications such as ChatGPT to be user-friendly, clear, and easy to interact with, their

chances to use them will rise significantly.

This discovery is a fundamental pillar of technology acceptance models, and it is strongly supported by the

recent empirical research on the use of AI in education. In his research on the acceptability of ChatGPT to

students, Strzelecki (2023) discovered that the expectancy of effort was a significant predictor of the intention

to use the technology and actual use. He has conducted research that underlines the fact that low learning curve

and the intuitive interface of such tools are critical in their quick acceptance of the tools by the students into

their workflow.

Similarly, Holzmann et al. (2025) investigated the generative AI continuous use determinants in students. Their

results emphasized that perceived ease of use was a key to the continuation of engagement, meaning that when

students do not think that the technology is easy to use, the initial interest in using it might not be converted

into sustained adoption. These studies interconnecting with the current findings serve to emphasize the

importance of reducing perceived complexity as a key to promoting the popular and long-term adoption of AI

tools in the academic environment.

Relationship between Social Influence and AI Adoption Behavior

The findings support the existence of a positive substantial relationship between social influence and the

adoption behavior of AI. This shows that the attitudes and actions of key referent groups, including peers,

lecturers, and the university community at large have a significant influence on whether students will use AI or

not. The social environment is an important source of normative pressure and information, which will lessen

uncertainty about a new technology (Tajfel & Turner, 1979).

Current literature confirms this result highly. In an international exploratory study, Faraon et al. (2025)

revealed that social influence was a key predictor of the intention of students to use ChatGPT in various

countries. They observed that the students tend to embrace AI when they perceive that individuals who are

significant to them ought to use it.

Moreover, another study by Cheng et al. (2022) on AI implementation, albeit in the healthcare environment,

offers a solid theoretical support. They showed that social influence, which can be provided by peers and

superiors, has a significant positive effect on the intention to adopt AI-assisted systems, which indicates the

universal impact of social networks on technological diffusion. As such, this paper confirms that the utility and

ease assessment of adopting AI is not only an individual, but it is also firmly rooted in a social environment in

which the perception and support of one group are a significant factor.

This study acknowledges key limitations that is, its findings are not broadly generalizable due to a

geographically restricted sample from one Malaysian university and a convenience sampling method that may

introduce bias. The reliance on self-reported survey data also posed another limitations that it created risks on

social desirability bias, thus, potentially inflating reports of AI adoption and positive perceptions. Furthermore,

the cross-sectional design captures only a snapshot in time, preventing causal conclusions or insight into

evolving behaviors. Finally, the research focused only on core facilitators from the UTAUT model. In this

manner, it omits potential barriers like academic integrity concerns, data privacy, and over-reliance, which

future studies should incorporate for a more balanced understanding.

Page 4828

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

In order to overcome the limitations of the study, future research should adopt longitudinal designs to track

evolving AI usage and establish causality. At the same time the research should integrate mixed methods to

enrich self-reported data with qualitative depth. In addition, the theoretical model must be expanded to include

factors like perceived risks, institutional support, and ethical considerations where other analysis techniques

like SEM is used. Finally, studies should be replicated across diverse institutions, disciplines, and cultural

contexts to significantly improve the generalizability and robustness of findings on AI adoption in higher

education.

To sum up, this research paper has verified that the usage of AI tools by the ABC university population is

largely determined by Performance Expectancy, Effort Expectancy, and Social Influence, as the UTAUT model

proposes (Li et al., 2023). The high positive correlations of all three hypotheses highlight the fact that the

students tend to adopt AI more when they find it useful, easy to use, and socially acceptable.

The results provide theoretical support of the UTAUT model within a modern educational setting and practical

recommendations on how universities should be guided. In the case of ABC university and other similar

universities, it is important to show the academic potential of AI strategically, make it easy to use, and create a

friendly social atmosphere to encourage responsible use. Although this work is limited due to its scope and

methodology, it offers a critical framework of future studies and practicable action on the changing AI situation

in higher education (Su et al., 2025). These recommendations must align with evolving ethical standards and

institutional policies to ensure responsible integration of AI.

ACKNOWLEDGEMENTS

We would like to express our appreciation to UNITAR International University for the publication fund in this

research.

Conflict Of Interest

The authors declare no conflicts of interest.

REFERENCES

1. Alessandro Rovetta, Bortolotti, A., & Palumbo, R. (2025). Integrating Team And Organizational

Identity: A Systematic

Literature Analysis.

Frontiers

In

Organizational Psychology,

2.

Https://Doi.Org/10.3389/Forgp.2024.1439269

2. Andrade, C. (2021). The Inconvenient Truth About Convenience And Purposive Samples. Indian

Journal Of Psychological Medicine, 43(1), 86–88. Https://Doi.Org/10.1177/0253717620977000

3. Black, R. W., & Tomlinson, B. (2025). University Students Describe How They Adopt AI For Writing

And

Research

In

A

General

Education

Course.

Scientific

Reports,

15(1),

1–10.

Https://Doi.Org/10.1038/S41598-025-92937-2

4. Bloomfield, R., & Rushby, J. (2024). Assurance Of AI Systems From A Dependability Perspective.

Arxiv (Cornell University). Https://Doi.Org/10.48550/Arxiv.2407.13948

5. Cheng, M., Li, X., & Xu, J. (2022). Promoting Healthcare Workers’ Adoption Intention Of Artificial-

Intelligence-Assisted Diagnosis And Treatment: The Chain Mediation Of Social Influence And

Human–Computer Trust. International Journal Of Environmental Research And Public Health, 19(20),

13311. Https://Doi.Org/10.3390/Ijerph192013311

6. Chong, L., Zhang, G., Goucher-Lambert, K., Kotovsky, K., & Cagan, J. (2022). Human Confidence In

Artificial Intelligence And In Themselves: The Evolution And Impact Of Confidence On Adoption Of

AI Advice. Computers In Human Behavior, 127. Https://Doi.Org/10.1016/J.Chb.2021.107018

7. Cornell University. (2023, August 16). AI & Academic Integrity | Center For Teaching Innovation.

Teaching.Cornell.Edu.

Integrity

Https://Teaching.Cornell.Edu/Generative-Artificial-Intelligence/Ai-Academic-

8. Cronbach, L. J. (1951). Coefficient Alpha And The Internal Structure Of Tests. Psychometrika, 16(3),

297–334. Https://Doi.Org/10.1007/Bf02310555

Page 4829

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

9. Cunningham, M.S. (1967) The Major Dimensions Of Perceived Risk: Risk Taking And Information

Handling In Consumer Behavior, Graduate School Of Business Administration, Harvard University,

Boston.

10. Davis, F. D. (1989). Perceived Usefulness, Perceived Ease Of Use, And User Acceptance Of

Information Technology. MIS Quarterly, 13(3), 319–340. Https://Doi.Org/10.2307/249008

11. Du, L., & Beibei Lv. (2024). Factors Influencing Students’ Acceptance And Use Generative Artificial

Intelligence In Elementary Education: An Expansion Of The UTAUT Model. Education And

Information Technologies. Https://Doi.Org/10.1007/S10639-024-12835-4

12. Faraon, M., Rönkkö, K., Milrad, M., & Tsui, E. (2025). International Perspectives On Artificial

Intelligence In Higher Education: An Explorative Study Of Students’ Intention To Use Chatgpt Across

The

Https://Doi.Org/10.1007/S10639-025-13492-X

13. Granić, A. (2023). Technology Acceptance

Https://Doi.Org/10.1007/978-981-19-2080-6_11

Nordic

Countries

And

The

USA.

Education

And

Information

Technologies.

And Adoption

In

Education.

183–197.

14. Guassi Moreira, J. F., Tashjian, S. M., Galván, A., & Silvers, J. A. (2021). Computational And

Motivational Mechanisms Of Human Social Decision Making Involving Close Others. Journal Of

Experimental Social Psychology, 93, 104086. Https://Doi.Org/10.1016/J.Jesp.2020.104086

15. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate Data Analysis (8th Ed.).

Cengage Learning

16. Hair, J. F., Hult, G. T. M., Ringle, C. M., Sarstedt, M., Danks, N. P., & Ray, S. (2021). Partial Least

Squares Structural Equation Modeling (PLS-SEM) Using R. In Classroom Companion: Business.

Springer International Publishing. Https://Doi.Org/10.1007/978-3-030-80519-7

17. Hansaram, S.K., Munap, R. (2025). Intersectionality at work: How disability, employer biases, and

SME constraints shape employment outcomes for persons with disabilities. Environment and Social

Psychology, 10(9), 4013

18. He, C., Shi, L., Yu, M., Jiang, Y., & Liao, C. (2024). Research On Influencing Factors Of Information

Adoption

Behavior

Of

College

Students

In

Cloud

Class.

601–606.

Https://Doi.Org/10.1145/3700297.3700401

19. Ho Ngoc Hai. (2023). Chatgpt: The Evolution Of Natural Language Processing. Authorea (Authorea).

Https://Doi.Org/10.22541/Au.167935454.46075854/V1

20. Hoo, W. C., Ching, K. Y. P., Cheng, A. Y., Saeed, K., & Shaznie, A. (2023). An examination on the

factors that influence the intention to use chatbots in Malaysia. International Journal of Management

and Sustainability, 12(3), 380–390. https://doi.org/10.18488/11.v12i3.3457

21. Holmes, W., Porayska-Pomsta, K., Holstein, K., Sutherland, E., Baker, T., Shum, S. B., Santos, O. C.,

Rodrigo, M. T., Cukurova, M., Bittencourt, I. I., & Koedinger, K. R. (2021). Ethics Of AI In Education:

Towards A Community-Wide Framework. International Journal Of Artificial Intelligence In Education,

32(1), 504–526. Https://Doi.Org/10.1007/S40593-021-00239-1

22. Holzmann, P., Gregori, P., & Schwarz, E. J. (2025). Students’ Little Helper: Investigating Continuous-

Use Determinants Of Generative AI And Ethical Judgment. Education And Information Technologies.

Https://Doi.Org/10.1007/S10639-025-13708-0

23. Hunziker, S., & Blankenagel, M. (2024). Cross-Sectional Research Design. Springer Ebooks, 187–199.

Https://Doi.Org/10.1007/978-3-658-42739-9_10

24. Izham, H. I. B., Peng, K. P. Y., Cheng, A. Y., Loy, C. K., & Hansaram, S. K. The Impact of Buy Now,

Pay Later Services on the Impulsive Buying Behavior of Generation Z in Shah Alam, Malaysia.

25. Jain, R., Garg, N., & Khera, S. N. (2022). Adoption Of AI-Enabled Tools In Social Development

Organizations In India: An Extension Of UTAUT Model. Frontiers In Psychology, 13.

Https://Doi.Org/10.3389/Fpsyg.2022.893691

26. Kiat, L. S., Hoo, W. C., Cheng, A. Y., Prompanyo, M., & Hossain, S. F. A. (2025). Factors Influencing

Intention to use 5G Mobile Technology and Adoption Onwards in Malaysia.

27. Kraatz, K., & Xie, S. (2023). Why AI Art Is Not Art – A Heideggerian Critique. Synthesis Philosophica,

38(2), 235–253. Https://Doi.Org/10.21464/Sp38201

28. Li, X., Zhao, Y., & Chen, Z. (2023). Perceived Ease Of Use And AI Adoption: Evidence From Higher

Education. Computers & Education, 180, 104567. Https://Doi.Org/10.1016/J.Compedu.2022.104567

Page 4830

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

29. Mat Roni, S., & Djajadikerta, H. G. (2021). Data Analysis With SPSS For Survey-Based Research.

Springer Singapore. Https://Doi.Org/10.1007/978-981-16-0193-4

30. Memon, M. A., Ting, H., Cheah, J.-H., Thurasamy, R., Chuah, F., & Cham, T. H. (2020). Sample Size

For Survey Research: Review And Recommendations. Journal Of Applied Structural Equation

Modeling, 4(2), 1–20. Https://Doi.Org/10.47263/JASEM.4(2)01

31. Merz, A., Moser, I., & Bergamin, P. B. (2025). Performance Expectancy And Social Influence Drive

The Acceptance Of Immersive Virtual Reality For Professional Collaboration. Virtual Reality, 29(3).

Https://Doi.Org/10.1007/S10055-025-01182-5

32. Millum, J., & Bromwich, D. (2021). Informed Consent: What Must Be Disclosed And What Must Be

Understood? The American Journal Of Bioethics, 21(5), 1–19.

33. Https://Doi.Org/10.1080/15265161.2020.1863511

34. Mohd Rokeman, N. R. (2024). Likert Measurement Scale In Education And Social Sciences: Explored

And

Explained.

EDUCATUM

Journal

Of

Social

Sciences,

10(1),

77–88.

Https://Doi.Org/10.37134/Ejoss.Vol10.1.7.2024

35. Moradi, H. (2025). Integrating AI In Higher Education: Factors Influencing Chatgpt Acceptance

Among Chinese University EFL Students. International Journal Of Educational Technology In Higher

Education, 22(1). Https://Doi.Org/10.1186/S41239-025-00530-4

36. Mumford, M. D., Higgs, C., & Gujar, Y. (2021). Ethics In Coercive Environments: Ensuring Voluntary

Participation In Research. Handbook Of Research Ethics In Psychological Science., 113–123.

Https://Doi.Org/10.1037/0000258-008

37. Mustafa, A. S., & Garcia, M. B. (2021). Theories Integrated With Technology Acceptance Model

(TAM) In Online Learning Acceptance And Continuance Intention: A Systematic Review. 2021 1st

Conference

On

Online

Teaching

For

Mobile

Education

(OT4ME).

Https://Doi.Org/10.1109/Ot4me53559.2021.9638934

38. Nazari, N., Shabbir, M. S., & Setiawan, R. (2021). Application Of Artificial Intelligence Powered

Digital Writing Assistant In Higher Education: Randomized Controlled Trial. Heliyon, 7(5), E07014.

Https://Doi.Org/10.1016/J.Heliyon.2021.E07014

39. Nhu, T., Nam Van Lai, & Quyet Thi Nguyen. (2024). Artificial Intelligence (AI) In Education: A Case

Study On Chatgpt’s Influence On Student Learning Behaviors. Educational Process: International

Journal, 13(2). Https://Doi.Org/10.22521/Edupij.2024.132.7

40. Openai. (2024). Openai Charter. Openai. Https://Openai.Com/Charter/

41. Paek, S., & Kim, N. (2021). Analysis Of Worldwide Research Trends On The Impact Of Artificial

Intelligence In Education. Sustainability, 13(14), 7941. Https://Doi.Org/10.3390/Su13147941

42. Pallant, J. (2020). SPSS Survival Manual: A Step By Step Guide To Data Analysis Using IBM SPSS

(7th Ed.). Routledge. Https://Doi.Org/10.4324/9781003117452

43. Păvăloaia, V.-D., & Necula, S.-C. (2023). Artificial Intelligence As A Disruptive Technology—A

Systematic Literature Review. Electronics, 12(5). Mdpi. Https://Doi.Org/10.3390/Electronics12051102

44. Rahman, M. (2023). Sample Size Determination For Survey Research And Non-Probability Sampling

Techniques: A Review And Set Of Recommendations | Journal Of Entrepreneurship, Business And

Economics. Www.Scientificia.Com . Https://Www.Scientificia.Com/Index.Php/JEBE/Article/View/201

45. Ruano-Borbalan, J.-C. (2025). The Transformative Impact Of Artificial Intelligence On Higher

Education: A Critical Reflection On Current Trends And Futures Directions. International Journal Of

Chinese Education, 14(1). Https://Doi.Org/10.1177/2212585x251319364

46. Ruslan, W. N. S. W. (2024). Exploring Drivers Influencing E-Commerce AI Adoption Among Social

Media

Natives.

Pakistan

Journal

Of

Life

And

Social

Sciences

(PJLSS),

22(2).

Https://Doi.Org/10.57239/Pjlss-2024-22.2.001069

47. Russell, S., Norvig, P., Fabrice Popineau, Laurent Miclet, & Cadet, C. (2021). Intelligence Artificielle :

Une Approche Moderne (4e Édition). Hal.Science. Https://Hal.Science/Hal-04245057

48. Safdar, M., Siddique, N., Gulzar, A., Yasin, H., & Khan, A. (2024). Does Chatgpt Generate Fake

Results? Challenges In Retrieving Content Through Chatgpt. Digital Library Perspectives.

Https://Doi.Org/10.1108/Dlp-01-2024-0006

49. Sánchez-Prieto, J. C., Cruz-Benito, J., Therón, R., & García-Peñalvo, F. (2020). Assessed By Machines:

Development Of A TAM-Based Tool To Measure AI-Based Assessment Acceptance Among Students.

International Journal Of Interactive Multimedia And Artificial Intelligence, 6(4), 80.

Page 4831

ww w .rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

50. Sewandono, R. E., Thoyib, A., Hadiwidjojo, D., & Rofiq, A. (2022). Performance Expectancy Of E-

Learning On Higher Institutions Of Education Under Uncertain Conditions: Indonesia Context.

Education And Information Technologies, 28(4). Https://Doi.Org/10.1007/S10639-022-11074-9

51. Shapiro, S. S., & Wilk, M. B. (1965). An Analysis Of Variance Test For Normality (Complete

Samples). Biometrika, 52(3-4), 591–611. Https://Doi.Org/10.1093/Biomet/52.3-4.591

52. Spears, R. (2021). Social Influence And Group Identity. Annual Review Of Psychology, 72(1), 367–

390. Https://Doi.Org/10.1146/Annurev-Psych-070620-111818

53. Strzelecki, A. (2023). Students’ Acceptance Of Chatgpt In Higher Education: An Extended Unified

Theory Of Acceptance And Use Of Technology. Innovative Higher Education, 49, 223–245.

Https://Doi.Org/10.1007/S10755-023-09686-1

54. Su, J., Wang, Y., Liu, H., Zhang, Z., Wang, Z., & Li, Z. (2025). Investigating The Factors Influencing

Users’ Adoption Of Artificial Intelligence Health Assistants Based On An Extended UTAUT Model.

Scientific Reports, 15(1). Https://Doi.Org/10.1038/S41598-025-01897-0

55. Tajfel, H., & Turner, J. C. (1979). An Integrative Theory Of Intergroup Conflict In Austin WG &

Worchel S.(Eds.), The Social Psychology Of Intergroup Relations (Pp. 33–47). Monterey, CA:

Brooks/Cole. [Google Scholar].

56. Tran, V. D. (2020). The Relationship Among Product Risk, Perceived Satisfaction And Purchase

Intentions For Online Shopping. The Journal Of Asian Finance, Economics And Business, 7(6), 221–

231. Https://Doi.Org/10.13106/Jafeb.2020.Vol7.No6.221

57. Uzun, L. (2023). Chatgpt And Academic Integrity Concerns: Detecting Artificial Intelligence

Generated Content. 3(1), 45–54. Https://Www.Researchgate.Net/Publication/370299956 Chatgpt And

Academic Integrity Concerns

58. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information

technology: Toward a unified view. MIS quarterly, 425-478.

59. Vieriu, A. M., & Petrea, G. (2025). The Impact Of Artificial Intelligence (AI) On Students’ Academic

Development. Education Sciences, 15(3), 343. Https://Doi.Org/10.3390/Educsci15030343

60. Yakubu, M. N., David, N., & Abubakar, N. H. (2025). Students’ Behavioural Intention To Use Content

Generative AI For Learning And Research: A UTAUT Theoretical Perspective. Education And

Information Technologies. Https://Doi.Org/10.1007/S10639-025-13441-8

61. Wen, E. C. Y., Hoo, W. C., Lee, A., & Cheng, A. Y. (2023). Mobile Banking Application (App)

Adoption Behaviour Amongst Malaysian Consumers. WSEAS Transactions on Business and

Economics, 20, 759–769. https://doi.org/10.37394/23207.2023.20.70.

62. Zhang, L., Shao, Z., Chen, B., & Benitez, J. (2024). Unraveling Generative AI Adoption In Enterprise

Digital Platforms: The Moderating Role Of Internal And External Environments. IEEE Transactions On

Engineering Management, 1–15.

Https://Doi.Org/10.1109/Tem.2024.3513773

Page 4832

ww w .rsisinternational.org