Çevrimiçi pratiklerle desteklenen ortaokul öğrencilerinin fen bilimleri sınav sonuçlarını etkileyen faktörlerin incelenmesi: CHAID Yöntemi ile öğrenme analitiği çalışması

In this study, within the scope of Learning Analytics; in a determined Learning Management System, the performance of the first semester Science exam of 6491 students studying in the 5th, 6th, 7th and 8th grades among different private schools in Turkey and the variables that affect the exam performance are predicted. For the analysis of the study, Chi-square Automatic Interaction Detector (CHAID) classification technique, which is one of the decision tree methods, offered the best result among the other classification methods used. According to the results of the research, it was estimated that the most important variables related to students' exam performance were the completed tests numbers and question numbers.