A Decision Tree Approach for the Classification of Mistakes of Students Learning SQL, a case study about SELECT statements

Abstract

TH Köln provides a web-based e-learning platform edb4, where novices can do their first steps in SQL. The goal of this paper is to build a decision tree (manually) that classifies the novice's errors. To do so we logged data containing tasks, solutions, and wrong statements over seven months and got a table with 7533 rows as a training set. Each leaf node of the decision tree is a class of errors of similar type and generates an error message with feedback to help the user to solve the task. Interesting and surprising are the mistakes that SQL novices make. The result improves the first steps of learning SQL in a simple and personalized way and gives the teachers hints to improve their learning outputs.