Data: Survey on factors influencing willingness to use ChatGPT

Description

This study investigates the factors influencing the intention to use ChatGPT among Chinese college students. Based on integrating the Technology Acceptance Model (TAM) and the Innovation Diffusion Theory (IDT), perceived popularity and compatibility were introduced to extend the model. Questionnaires were distributed to 451 college students through snowball sampling, and the data were statistically analyzed using SPSS and AMOS software. The research results showed that perceived compatibility and perceived popularity have a significant positive effect on the intention to use ChatGPT; perceived usefulness and perceived ease of use played a partial mediating role between perceived compatibility, perceived popularity, and use intention; perceived usefulness played a partial mediating function between perceived ease of use and use intention. The study provides a theoretical perspective for understanding college students' acceptance and intention to use emerging technology tools like ChatGPT. The findings also provide an empirical basis for educators and technology developers to optimize product design and enhance the effectiveness of educational technology applications. In selecting subjects for data collection, this study draws on the research standards established by researchers at Harvard, MIT, and Stanford for classifying MOOC platform users, as well as the clustering criteria used by American scholar Khalil to categorize Coursera learners. ChatGPT users are defined as those who actively engage with the platform, excluding "no-shows" and "registrants only." Given the sample's availability and representativeness, the survey was conducted using a combination of online and offline methods. The offline survey was conducted from October to December 2023, during which research team members distributed questionnaires to college students by participating in academic forums, themed events, and seminars on college campuses. A total of 110 valid questionnaires were collected. The online survey was carried out from January to April 2024, primarily through social networks, WeChat, QQ, and other online channels connected to the offline survey participants, resulting in 382 returned questionnaires. In total, 492 questionnaires were collected through both methods. After excluding invalid responses due to excessively short or long completion times and incomplete information, 451 valid questionnaires were obtained, yielding an effective rate of 91.67%.

Steps to reproduce

In selecting subjects for data collection, this study draws on the research standards established by researchers at Harvard, MIT, and Stanford for classifying MOOC platform users, as well as the clustering criteria used by American scholar Khalil to categorize Coursera learners. ChatGPT users are defined as those who actively engage with the platform, excluding "no-shows" and "registrants only." Given the sample's availability and representativeness, the survey was conducted using a combination of online and offline methods. The offline survey was conducted from October to December 2023, during which research team members distributed questionnaires to college students by participating in academic forums, themed events, and seminars on college campuses. A total of 110 valid questionnaires were collected. The online survey was carried out from January to April 2024, primarily through social networks, WeChat, QQ, and other online channels connected to the offline survey participants, resulting in 382 returned questionnaires. In total, 492 questionnaires were collected through both methods. After excluding invalid responses due to excessively short or long completion times and incomplete information, 451 valid questionnaires were obtained, yielding an effective rate of 91.67%. As shown in Table 1, the survey sample distribution is relatively balanced and structurally reasonable in terms of gender, discipline, education level, and monthly usage frequency, providing reliable data support for subsequent statistical analyses. During data processing, SPSS 27.0 was used for descriptive statistical analysis of the scales, reliability testing, and common method bias testing. Additionally, AMOS 27.0 was employed for confirmatory factor analysis, validity testing, and structural equation modeling. All testing procedures strictly adhered to the technical guidelines and recommendations provided by Hair et al. in the relevant field.

Institutions

Categories

Funding

Licence