HappyFace: Identifying and predicting frustrating obstacles for learning programming at scale

摘要: Unnecessary obstacles limit learning in cognitively-complex domains such as computer programming. With a lack of appropriate feedback mechanisms, novice programmers can experience frustration and disengage from the experience. In large-scale educational settings, struggles learners are often invisible to infrastructure have limited ability seek help. this paper, we perform collection code snippets an online learn-to-code platform, Python Tutor, collect rating through light-weight learner mechanism. We then devise technique that automatically identify sources based on participants labeling their levels. found 3 factors best predicted programmers' state: syntax errors, using niche language features, understanding with high complexity. Additionally, evidence could predict frustration. Based these results, believe embedded mechanism lead future intervention systems.