New Strategies for Teaching Statistics at School

The idea that the teaching of statistics should emphasise the understanding of statistical concepts and methods, and also encourage the students to acquire awareness of how important it is to be able to assess the mass of statistical information they come across every day has a similar consensus. The achievement of these aims is facilitated only if students, in their learning experience, are asked to handle concrete problems based on real settings. In that respect, a relevant contribution has been given by David Moore (1997), who suggests that "the most effective learning takes place when content, pedagogy and technology reinforce each other in a balanced manner". This framework, however, rarely finds concrete application in schools, especially in our country. Here the teaching of statistics, besides being marginal and incorporated in the mathematical curriculum, is very often dealt in an abstract and formal way. There is very limited emphasis on the meaning of concepts and even less on the interpretation of the results. Students need to see mathematics and statistics in a correct perspective, and to distinguish the different features of their logic. In particular, it is fundamental to enable learners to appreciate the role of statistical reasoning, which allows them to read and interpret real phenomena with no pretension of giving "exact answers", nevertheless helping them to develop a more objective view of reality. The teaching experiment in our research aimed at creating a model for statistics introductory courses based on an approach uncommon in Italy, which we shall refer to as DOA (Data Oriented Approach). The DOA approach is based on the hypothesis that working with real data reflecting a real-life phenomenon favours a better learning of statistical methods and develops students'interests, as they are personally involved in the collection and interpretation of data. To verify the validity of this model and to identify the most suitable strategies for its implementation, we were able to rely on the willingness of the teachers involved in the study to experiment with different types of course management. The research was carried out in the four CIRDIS, units and involved 145 teachers and 2,129 pupils from primary schools (age 6-10), 86 teachers and 1514 pupils from lower secondary school (age 13-14) and 107 teachers and 2500 pupils from secondary school.