Evaluating Children’s Code Quality and Complexity in Computational Thinking Games

With a growing focus on science, technology, engineering, math and computer science (STEM-CS) for the jobs of the future, computational thinking has emerged as a key topic of interest in K-12 education. Children that are exposed at an early age to STEM curriculum, such as computer programming and computational thinking, demonstrate fewer obstacles entering technical fields. Coding is explicitly regarded as a key 21st Century Skill, as the “literacy of today,” such that the acquisition of digital competences, including coding, is regarded as essential to sustain economic development and competitiveness. Using a popular programming app, codeSpark Academy, this study evaluates the quality, complexity, and abstraction of code produced by children. We are currently in the process of analyzing log-data from 4000 student accounts (children ages 5-8) with game-play telemetry collected by codeSpark Academy’s computer science learning games platform over the years 2018 and 2019. We will present answers to the following research questions: How are code metrics evolving over time, per student and per game type? How do code metrics distinguish across different account types, game types and levels?