PALOMAR PROJECT - PREDICTING SCHOOL RENOUNCING DROPOUTS, USING THE ARTIFICIAL NEURAL NETWORKS AS A SUPPORT FOR EDUCATIONAL-POLICY DECISIONS

The ''Palomar'' project confronts two problem situations that are part ly independent and partly connected to the Italian schooling system: u nstable participation in school such as drop out and educational guida nce. Our concern is that of a set of phenomena which consists of ceasi ng compulsory education, repetition of a year at school, school ''drop outs'', irregular compulsory attendance and delays in the course of s tudies. The ''Palomar'' project is designed to offer educators and adm inistrators who want to effectively intervene with these complex probl ems to furnish school guidance services as an instrument able to: 1. P redict: creating a system able to predict in advance (not in a ''cause -effect'' way but as an approximation): a) which students are at ''ris k'' for school destabilization or failure; b) what are the prototypica l characteristics of these students; c) which students among those stu died are more likely to ''destabilize'' or fail in school; in which co urse of study does each student have the greatest chance of success; d ) which, among the variables studied and appropriately weighted for ea ch student, will predict the successful grade, analyzed for each possi ble course of studies. 2. Optimize: selecting and focusing on a studen t on the basis of the information given. It is possible: a) to point o ut which personal factors (relational, familial, student, disciplinary , economical) need to be reinforced in order to improve the school per formances of each selected student, both to prevent or limit ''droppin g out'' desertion or failure and to raise the performances in the chos en school course as much as possible; b) on the basis of what was ment ioned above, to simulate the possible support measures to increase the efficacy of the considered intervention; c) to choose for each studen t the appropriate intervention strategy capable of obtaining the maxim um result and the maximum efficacy in the given conditions. 3. Verify: when the strategy of intervention has been decided and we proceed wit h its implementation, it is possible to periodically verify (''follow- up''), through subsequent administration of the form, the outcome vari ations elapsed in the prediction of school success or failure. This ma kes it possible to verify in itinere the efficacy of the interventions carried out and, if necessary, to create variations and adjustments. 4. Produce scenarios: the application field of the Prediction System w ith Artificial Neural Networks can also be one of a group of students, of one or more organized units (for example a class, a school, or a g roup of schools). In this case the Prediction System ANN using the pro gram SCHEMA (Buscema, 1996b) is able to: a) determine intervention str ategies in order to optimize and to produce the maximum results of a g roup of students as the one of a class; b) optimize the formative rout e of a whole institute in order to prevent or limit the need for schoo l guidance.