The accuracy level of ChatGPT in information search by social science students

Mazli, Nurul Najihah (2025) The accuracy level of ChatGPT in information search by social science students. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Preview

Text (Full Text) The accuracy level of ChatGPT in information search by social science students.pdf - Submitted Version Download (1MB) | Preview

Abstract

Nowadays, artificial intelligence is widely used in every sector including in education. Most students especially university students, are very efficient in using AI tools like ChatGPT to help them do their work like writing assignments and essays and finding pieces of information. They already made the AI tool one of their reference materials. It is good as they have a lot of sources that students can refer to while doing their work. However, the students should not be too dependent on the information provided by the ChatGPT. This is because it is still a technology and the accuracy level also is not convincing. Therefore, this research aims to study the accuracy level of information retrieved by social science students using ChatGPT and determine the relationship between three independent variables (validity, reliability, and relevance) and dependent variables (information search by social science students). This study focused on the Universiti Teknikal Malaysia Melaka (UTeM) social science students. The data was collected from 215 respondents through the questionnaire survey. Therefore, the result from the Multiple Regression Analysis and Pearson’s Correlation Coefficient showed that accuracy levels which are validity, reliability, and relevance had a significant relationship and strong relationship to information search by social science students. To conclude, through this research, hope that students will be more careful with the information provided by ChatGPT and ensure that it is valid before using it.

Actions (login required)

View Item