Data Mining Techniques in Identifying Factors Associated with Schoolchildren Science and Arts Academic Achievement

The study aims to determine students' factors that affect and predict their academic achievements in Science and Arts. The study group consisted of 1945 students in grades 5-9, aged 11-16 years studying at public and refugee schools in West-Bank. The data were collected using the health behaviors school children questionnaire in the 2013-2014 academic year and analyzed using Chi-squared Automatic Interaction Detection (CHAID) data mining technique to discover the relationship between the factors associated with academic performance in Science and Arts. The results indicated 63% model accuracy in classifying and predicting the associated factors with schoolchildren's Arts and Science achievements. Two different decision trees were identified for Arts and Science using the CHAID data mining model. The Arts and Science achievements have a strong association with PTSD levels, positive health perception, age, and depression levels. Arts achievement has a higher association with life satisfaction, Psychosomatic symptoms, healthy food consumption, family support, and gender than Science achievements. Science achievement has a stronger association with unhealthy food consumption, school support, and school type. Classification and prediction revealed from the data mining method in our data set showed the potential of data mining methods in discovering new information and knowledge that might not be found in conventional analysis. The CHAID model seems to be an effective tool for identifying the associated factors that affect student achievements.