Geometry students, while moving from junior high-school to Lyceum, gradually cease to deal with practical topics and have to confront with other topics of theoretical nature. Worldwide research has shown that Lyceum students have great difficulty in writing formal proofs in geometry, particularly when traditional teaching methods are employed in the teaching of geometry. In this paper, we focus on the findings of a research project, which is part of a wider research that aims at investigating the ability of 15 years-old novice geometry students to write formal proofs. In particular, we use data from a random sample of schools in Athens from students attending the first year of Lyceum. Our research findings evidence that, while attempting to write formal geometry proofs, students who had employed a tool, called ``Reasoning Control Matrix for the Proving Process'' (RECOMPP) had significantly improved their ability in writing formal geometry proofs than those who had not employed this tool.