Prediction of emotional competence in future education teachers through a multiple logistic regression model and classification trees

Abstract

In order to know the level of emotional competence of future education teachers and evaluate the incidence that different personal and academic factors, we have used binary logistic regression and classification trees. Non-experimental design was carried out, with a sample of 359 students from the Faculties of Education of the University of Malaga and Almería. The results show that variables such as university access scores, the average mark of the first four-month period of the first year of the degree, the effort before the study, being a repeater, the choice of the degree as the first option and having siblings are significant predictors in the logistic model, where the last variable has the greatest determination capacity. Furthermore, in the segmentation tree it has been obtained that those students who do have siblings have a higher level of development of emotional competence, with a probability of 69.2% and if they have also chosen the degree as the first option, this percentage is increased to 75.9%. It can be concluded that it is essential to institutionally establish sufficient conditions so that teachers can detect the predictor variables and establish training strategies that allow increasing the level of development of the competence emotional.