A Decision Tree Approach to Predict the Online Information Search Performance of 8-13 Year Old Students in French-Speaking Belgium

In French-speaking Belgium, one of the new tasks of the teacher will be to develop the informational skills of learners. Therefore, the main purpose of this research is to identify the variables that affect the performance of online information retrieval so that teachers can plan appropriate teaching activities. The statistical Classification and Regression Tree (CART) technique was used to characterize the activity of 260 students aged 8 to 13 according to their results in three search tasks. The results show that certain navigational strategies, as well as the different individual characteristics of the participants in our study, can be considered as key factors that modulate performance. The different findings from our analyses allow us to put forward concrete avenues for the mastery of this skill in an educational context.