The analysis of e-learning settings, which are prepared on the basis of multiple intelligence domains determined by artificial intelligence in science instruction, as per different variables

This study aims to analyze whether e-learning settings that are prepared on the basis of multiple intelligence domains determined by artificial intelligence in science instruction have any effect on academic achievements, science attitudes, and technology attitudes of middle school students. The study was carried out within the framework of a multi-stage design, one of the mixed method designs. The study consists of three stages in total. In the first stage of the study, the data set to be used in the second stage was prepared. In the second phase of the study, the "design-based research method" was utilized to design, develop, use, and evaluate the website and report about it. This phase of the study was carried out with a total of 527 fifth-grade students (246 female students, 281 male students) enrolled in five different middle schools in Elâzığ province of Turkey in the 2021-2022 academic year. In the third phase of the study, the dominant intelligence types of students in the experimental group were predicted by the model created in the second phase of the study, and e-learning settings were created so that students in the experimental group would receive an education suitable to their predicted dominant intelligence types. The third phase of the study was carried out with the participation of a total of 130 students (58 female students, 72 male students) enrolled in a middle school in Elâzığ province in the 2022-2023 academic year and assigned to one experimental group and three control groups. The quantitative data used in the second phase of the study were collected with the Personal Information Form, the Video Ranking for Areas of Learning, and the Scale of Multiple Intelligence for Children whilst the qualitative data were collected via focus group interviews as well as interviews with teachers. The quantitative data used in the third phase of the study were collected with the Personal Information Form, the Video Ranking for Areas of Learning, the Academic Achievement Test, the Assessing Attitudes and Preferences in Science Instrument, and the Attitudes Toward Science and Technology Lesson Scale while the qualitative data were collected via interviews with students in the experimental group and observations on students in control groups. In the analysis of quantitative data, the ANCOVA was used with the SPSS 23.0 whilst the content analysis was utilized in the evaluation of qualitative data. The validity and reliability of the videos prepared in the first stage of the study were ensured in line with expert opinions. According to the results obtained in the second phase of the study, the model attaining the best accuracy was identified as the Extra Trees algorithm (with an accuracy rate of 97.24%). Next, according to the results of the ANCOVA, the academic achievement levels of the students in the experimental group had no statistically significant difference from those of the students in the control group 1 and the control group 3, on the other hand, students in the experimental group had higher academic achievement levels than students in the control group 2 and this difference between the two groups was statistically significant. Attitudes of students in the experimental group toward science course had no statistically significant difference from those of students in the three control groups whereas technology attitudes of students in the experimental group had statistically significant differences from those of students in the three control groups. Lastly, according to qualitative data of the study, students often stated that learning their dominant intelligence types via artificial intelligence was quite important to their educational lives, they would like to receive an education suitable to their AI-based dominant intelligence types if they had the option to do so, they could more easily understand the course unit instructed during the process, their interest in, enthusiasm for, and curiosity about technology increased, the practice provided them with several benefits, and such practices should not be limited only to a unit in science course but also be used in different courses.