Scratch Analysis Tool(SAT): A Modern Scratch Project Analysis Tool based on ANTLR to Assess Computational Thinking Skills

Abstract

With the introduction of computer programming in schools around the world, Scratch has risen in prominence for its thinkable, meaningful and social. Aiming to assessing the Computational Thinking skills of a Scratch program, we design and implement a new Scratch program analysis tool based on ANTLR. To solve some flaws (e.g., high failure rate and low efficiency) in Dr. Scratch which is the most relevant tool to assess Computational Thinking skills of Scratch programs, we choose the recognition tool ANTLR to design the system module and the assessing flow. And then, we customize more than 200 lexical and syntax parser rules in ANTLR. Furthermore, we expand the grading standard of assessing Computational Thinking skills in Dr. Scratch. Some fundamental concepts in Computer Science, such as stack, queue and recursion method, are involved in our grading standard. Experiment results show the performance (e.g., success rate, execution time) of SAT is superior to that of Dr. Scratch.