Selection method of variables from a set of dyslexia screening tasks in french school age children

Dyslexia is a specific disorder of language. Researches led on dyslexia origin have conducted to multiple hypotheses and various rehabilitation treatments. In order to help in dyslexia diagnosis, a preliminary test was created. It includes the most representative dyslexia screening tasks. But, this test is too long and has too many variables. So, to define the best combination of tasks and reduce the number of variables, a method using PCA has been developed to select a first set of variables. Then, two methods (stepwise discriminant analysis and logistic regression) were applied on the pre-selected variables to identify dyslexia detection models.