PISA 2018’de Okuduğunu Anlama Başarısını Yordayan Değişkenlerin Veri Madenciliği İle Belirlenmesi
Determining Variables That Predict Reading Comprehension Success by Data Mining in PISA 2018
Author(s): Yusuf Kasap, Nuri Dogan, Cem KOÇAKSubject(s): Higher Education , Methodology and research technology
Published by: Celal Bayar Üniversitesi Sosyal Bilimler Enstitüsü
Keywords: Reading comprehension; Data mining; Classification; Estimation; PISA;
Summary/Abstract: In this study, it was aimed to determine the important variables that can predict the success of reading comprehension with 34 independent variables obtained from the student questionnaire given to the students who participated in PISA in 2018. For this purpose, 79 countries participating in PISA were put in order of success percentage. Then, with equal slices, countries were divided into lower, middle and upper slices. The study sample was created by selecting 9 countries including Turkey according to their percentiles. In the analyzes made with the logistic model using Turkey, low, medium, high achievement group and study sample data, the important variables predicting reading comprehension success varied between 5 and 8. According to the results obtained, for Turkey, low, middle and high achievement group countries and study sample, it is determined that the important common predictors of success are, the difficulty perception of the PISA test, the highest education level index of the parents and the educational items in the house. Then, classification, cross-validation and prediction performances of success with important variables were calculated. It has been observed that with the estimations made with important variables, results were close to the performances of the models obtained with 34 variables.
Journal: Celal Bayar Üniversitesi Sosyal Bilimler Dergisi
- Issue Year: 19/2021
- Issue No: 04
- Page Range: 241-258
- Page Count: 18
- Language: Turkish