ChatGPT and AI in learning intention questionnaire (UTAUT2)

Students using ChatGPT and other generative AI tools to study is a new topic that many faculties care about. UTAUT2 of Venkatesh, Thong and Xu (2012) fits because use is voluntary and involves enjoyment and habit. The template drops price value because most students use the free version.

Research model

Model following UTAUT2 (Venkatesh et al., 2012), without price value: performance expectancy, effort expectancy, social influence, facilitating conditions, hedonic motivation and habit affect behavioral intention.

• Performance Expectancy → Behavioral Intention (positive)
• Effort Expectancy → Behavioral Intention (positive)
• Social Influence → Behavioral Intention (positive)
• Facilitating Conditions → Behavioral Intention (positive)
• Hedonic Motivation → Behavioral Intention (positive)
• Habit → Behavioral Intention (positive)

Questionnaire items

Original English wording. Answered on a 5-point Likert scale (1 = Strongly disagree … 5 = Strongly agree) unless the scale page says otherwise; (R) marks a reverse-coded item.

Performance Expectancy (PE) · UTAUT2 scale

Code	Item
PE1	I find [mobile Internet] useful in my daily life.
PE3	Using [mobile Internet] helps me accomplish things more quickly.
PE4	Using [mobile Internet] increases my productivity.

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Effort Expectancy (EE) · UTAUT2 scale

Code	Item
EE1	Learning how to use [mobile Internet] is easy for me.
EE2	My interaction with [mobile Internet] is clear and understandable.
EE3	I find [mobile Internet] easy to use.
EE4	It is easy for me to become skillful at using [mobile Internet].

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Social Influence (SI) · UTAUT2 scale

Code	Item
SI1	People who are important to me think that I should use [mobile Internet].
SI2	People who influence my behavior think that I should use [mobile Internet].
SI3	People whose opinions that I value prefer that I use [mobile Internet].

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Facilitating Conditions (FC) · UTAUT2 scale

Code	Item
FC1	I have the resources necessary to use [mobile Internet].
FC2	I have the knowledge necessary to use [mobile Internet].
FC3	[mobile Internet] is compatible with other technologies I use.
FC4	I can get help from others when I have difficulties using [mobile Internet].

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Hedonic Motivation (HM) · UTAUT2 scale

Code	Item
HM1	Using [mobile Internet] is fun.
HM2	Using [mobile Internet] is enjoyable.
HM3	Using [mobile Internet] is very entertaining.

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Habit (HT) · UTAUT2 scale

Code	Item
HT1	The use of [mobile Internet] has become a habit for me.
HT2	I am addicted to using [mobile Internet].
HT3	I must use [mobile Internet].

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Behavioral Intention (BI) · UTAUT2 scale

Code	Item
BI1	I intend to continue using [mobile Internet] in the future.
BI2	I will always try to use [mobile Internet] in my daily life.
BI3	I plan to continue to use [mobile Internet] frequently.

Source: Venkatesh et al. (2012). See UTAUT2 scale.

Tips

• Explain in Chapter 2 why price value is dropped; if your sample has many paying users of the premium version, keep PV.
• The original performance expectancy items ask about daily life and productivity; this template rewrites them for studying and keeps the original items for comparison.
• Intention to use AI for study is easily affected by instructor and university rules; consider asking about these rules and ensure anonymity so students answer honestly.
• A Vietnamese study using UTAUT2 for ChatGPT had some constructs with Cronbach's alpha below 0.70; run a pilot with 30-50 students before the main survey.

Work out the minimum sample size for this model; once the data are in, clean the responses and run the survey data analysis for your results chapter.

Studies that used this model

• Duong, Bui, Pham, Vu and Nguyen (2023). How effort expectancy and performance expectancy interact to trigger higher education students' uses of ChatGPT for learning. Interactive Technology and Smart Education, 21(3), 356–380. Sample: 1,461 university students in Vietnam; UTAUT; polynomial regression and response surface analysis. Reliability: .
• Nguyễn Thị Thanh Thúy et al. (2025). Các yếu tố ảnh hưởng đến ý định sử dụng ChatGPT trong học tập và nghiên cứu của sinh viên Đại học Đà Nẵng. Tạp chí Khoa học Kinh tế, Trường Đại học Kinh tế – Đại học Đà Nẵng, 11(04), 102–117. Sample: 300 students at the University of Da Nang; UTAUT2; regression. Reliability: α 0.608-0.804 (Table 3), a few constructs below 0.70.

Collecting responses

If your own network cannot reach the target respondents, you can pay for verified responses through Genius Data (referral link, we may earn a commission), paying only for approved responses.