A study of data from families involved in the PICAPS project was conducted. The main objective of the survey was to understand the characteristics of school-age children who do not attend school. The information collected concerned both personal data - composition of the household, age, sex - and socio-economic data - schooling of family members, income, work activities. A first part of the analysis was purely descriptive, with the aim of defining more precisely the characteristics of out-of-school children. The second part of the analysis sought to identify which characteristics of the household were most influential in the dropout phenomenon, in order to construct a general model. For this purpose, a classification algorithm was trained to provide probability rates of dropout based on household characteristics. The final result of this analysis is a set of classification rules that can be used to prevent dropout through the planning of more targeted interventions of socio-educational support.
Further information at with the full technical report