Exploring self-learning efficacy: a comparative analysis of ChatGPT and Gemini
Purpose
ChatGPT and Gemini stand out as prominent artificial intelligence (AI) platforms with a substantial user base. These platforms are being widely used across different domains owing to their versatility. Learning is an integral part of human development. So far, no studies have been conducted to explore the use of AI in this domain, particularly, self-learning. The purpose of this study is to investigate Gemini and ChatGPT’s usability in the context of self-learning. This study also seeks to compare the effectiveness of these two systems’ self-learning features.
Design/methodology/approach
This study involves a comparative assessment of two Generative AIbots: ChatGPT and Gemini. Three concepts were selected to be learnt in the realms of Science, Humanities and Literature. A pre-test post-test two group experimental design was selected. The sample consisted of 80 pre-service teachers. The pre-test and post-test scores were analyzed using SPSS 23. Homogeneity of both the groups was obtained through Levene’s test. Group “A” was exposed to concepts explained by ChatGPT, while Group “B” received the same explanations via Gemini. Independent t-test for the post-test scores of both groups was run to check which was a better platform.
Findings
Both ChatGPT and Gemini are good platforms for self-learning. On conducting an independent t-test for the post-test scores of both groups, it became evident that those who learned through ChatGPT exhibited superior performance. Hence, it can be concluded that ChatGPT is a better platform for self-learning as compared to Gemini.
Research limitations/implications
This study was conducted only on pre-service teachers and may not reflect the learning effectiveness of the AI chatbots on other groups of learners like school students or other working professionals. Although Levene’s test was performed based on pre-test scores to ensure homogeneity in terms of knowledge between the groups, factors such as IQ and prior knowledge of the pre-service teachers were not accounted for. This study focuses on only three concepts from Science, Humanities and Literature. The results of this study might vary with different subjects or more complex concepts.
Practical implications
At higher education levels, most of the learning is self-directed. There are many platforms through which learning can take place. This study gives direction as to which platform is better and the probable reasons for it. There are more platforms that can also be studied as well as re-study can be done, as AI platforms are constantly being updated.
Social implications
This study would help a majority of learns in the higher educational level and those who depend on AI form learning.
Originality/value
No comparative experimental study has been done between ChatGPT and Gemini with regard to self-learning.
