On the Students’ Misconceptions in Object-Oriented Language Constructs

Analyze the Object-oriented (OO) source code developed by students provides useful formative tips to instructors. According to this, it is essential to understand the student’s real difficulties allowing instructors to shape effective courses. To provide run-time feedback to students and to study and analyze the evolution of their performances offline and over time we designed a framework and developed a tool. It allows to identify students’ misconceptions analysing source code and to create personalized student reports automatically. In this paper, we present an empirical study, conducted using our toolchain, that involves 1627 projects extracted from the multi-institution Blackbox dataset. We identified a violation model for Java language constructs based on established results in the computing education community. Afterwards, we grouped such violations in categories and analyzed the relations among them. Our contributions might be helpful in delivering formative feedback and supporting instructors who teach Java and object-oriented programming in general.