The purpose of the present study is to examine the properties of a curriculum-based pre-referral screening tool. According to the current Greek preschool curriculum, preschoolers' participation is evaluated across five core axes, namely Play, Routines, Daily Activities, Explorations, and Organized Activities. Teacher evaluations sorted 201 preschoolers into one of three groups: (1) referred to a pre-referral intervention program, (2) referred to an educational diagnosis process, or (3) not referred at all. Classification And Regression Tree (CART) analyses revealed that teachers' ratings on the tool can correctly predict the pupils' membership to both of the pre-referral and not referral groups. In fact, depending on the targeted group, prediction accuracy ranged as high as 85% to 100%. The inclusion of participation in activities requiring fine motor skills and body relaxation increased the prediction rates to 97.5% for both groups.