Dataset on the perception and intention of students toward AI-powered virtual tutors for digital English learning
Description
This data assesses the determinants influencing students’ intention to utilize AI-powered virtual tutors for digital English education. The investigation is grounded in a theoretical framework proposing that six specific features of virtual tutors, aesthetics, anthropomorphism, availability, empathy, enjoyment, and learning content, affect users’ behavioral intentions through the dual mediating roles of perceived trust and perceived value. A quantitative research methodology was adopted, utilizing a structured online questionnaire distributed to high school and university students in Vietnam. The study implemented a non-probability convenience sampling technique to gather information from the target demographic. The survey tool was designed to record demographic profiles alongside respondents’ perceptions regarding the nine established theoretical constructs. The data collection process resulted in 645 valid responses, organized into a dataset containing 40 columns, where the initial four columns detail demographic characteristics and the subsequent 36 columns represent the individual measurement items for the variables. The provided file, named AI-powered.virtual.tutors.for.digital.English.learning_dataset, delivers detailed insights into how specific technological attributes influence learners’ cognitive and emotional assessments. The analytical outcomes indicate that the majority of the hypothesized AI tutor features significantly enhance both perceived trust and perceived value. Notably, the data reveals one exception: anthropomorphism did not demonstrate a statistically significant effect on the perceived value of the tutor. Furthermore, the statistical results confirm that both perceived trust and perceived value act as robust predictors that strongly drive the students’ ultimate intention to use these digital learning systems. These findings present empirical guidance for educational technology developers and instructional designers to optimize virtual tutoring platforms. By prioritizing elements that actively build trust and educational value, such as high-quality learning content, system availability, and empathetic interactions, creators can more effectively foster student adoption. Additionally, this dataset functions as a valuable asset for exploring the technological drivers of AI acceptance in modern education. Scholars can utilize this information to test extended educational technology acceptance models, conduct comparative analyses between different educational tiers (high school versus university), and develop targeted strategies for implementing AI assistance in language acquisition programs.
Files
Institutions
- FPT UniversityHanoi, Hanoi