Resilient students in mathematics based on PISA 2003 and 2012 surveys for Greece: secondary analysis using classification trees (Master thesis)

Καλαϊτζίδου, Μαγδαληνή/ Kalaitzidou, Magdalini

Nowadays, there is increasing interest in the international literature on the charac-teristics of students who exhibit resilience. Resilient students are the students who, despite the social and economic difficulties, have managed to have satisfactory performance in school. The purpose of this study is to investigate the characteristics of resilient students in mathematics through the PISA survey data of 2003 and 2012 for Greece. The PISA study is an international assessment measuring performance of students at the end of their compulsory education, in three school subjects, Mathematics, Natural Sciences and Text Comprehension, and is held every three years. In particular, this study investigates the social and educational characteristics of resilient students, their attitudes towards mathematics and ICT, their contact with ICT, their perceptions for their class teachers and the learning strategies they say that adopt in mathematics, as well as their relationship with resilience to mathematics. Data analysis was based on Classification Trees, a modern method of the growing field of Machine Learning. Classification Trees are used for classification or clustering of subjects or objects based on a set of independent variables, which are related with each other and act on a basic dependent variable in a nominal/ordinal scale. The final predictive model is represented with a graph having a tree shape. The application of Classification Trees on the 2003 and 2012 PISA data for Greece revealed that student resilience in mathematics depends on the time spent by students for dealing with Mathematics and Science per week, their familiarity with mathematical concepts, their interest in mathematics, math self-efficacy and self-concept, math anxiety and the student ability to solve complex problems. In conclusion, resilience to mathematics can be affected by social and educational characteristics of students, their attitudes towards mathematics and the adopted learning strategies. Finally, these findings can be used to inform educational policies so as to promote strategies that improve student resilience
Institution and School/Department of submitter: Δημοκρίτειο Πανεπιστήμιο Θράκης. Σχολή Επιστημών Αγωγής. Παιδαγωγικό Τμήμα Δημοτικής Εκπαίδευσης
Subject classification: Education--Research
Mathematics--Study and teaching (Elementary)--Evaluation
Keywords: Ανθεκτικοί μαθητές,Έρευνα PISA,Μηχανική μάθηση,Δέντρα ταξινόμησης,Resilient students,Machine learning,Classification trees

Files in This Item:
File Description SizeFormat 
KalaitzidouMagdalini .pdfΜεταπτυχιακή εργασία2.87 MBAdobe PDFView/Open    Request a copy

 Please use this identifier to cite or link to this item:
  This item is a favorite for 0 people.

This item is licensed under a Creative Commons License Creative Commons