Abstract:

With this study we propose to investigate exploratory models about the effects of factors – students’ gender, age and school level – that may interfere with attitudes, affect and emotions toward learning, while explicitly modelling the multilevel structure of a large data set from an educational context. Based on a study with Portuguese youngsters (N=778) from regular classes, whom we applied the QATS – Questionnaire of Attitudes Toward School and the Emotional Intelligence Inventory (Candeias, in press, Candeias, 1997), we use regression trees algorithm to predict student’ attitudes, affect and emotions toward learning. Predictors include age, gender as personal attributes and some specific characteristics as school level.
The results come to reveal the importance of understand attitudes toward learning, as being affected for different combinations of factors. Attitudes should be understood based on the natures of the subject. Thus the improvement of attitudes and motivation toward school should involve students, teachers, parents and community.