This interesting paper reports on a tutoring system designed to help undergraduate students taking a database course that uses structured query language (SQL) for query operations. The major thrust of the automated tutoring system is training rather than knowledge acquisition.

Students attempt several tasks, by way of answering questions presented to them by the system during the interactive, learn-by-doing sessions. The outcome of each of these tasks could be one of success, partial solution, or a specific category of error, or an outcome that requires the tutoring system to generate appropriate feedback to the user. The tutoring system tracks this information dynamically, and creates a personalized guidance scheme for each student throughout a tutoring session. Guidance messages (“You are having a small number of problems with selecting correct attributes,” or “You are having a lot of problems with aggregate functions”) create a virtual classroom scenario, wherein a simulated instructor renders advice based on the past performance of the student.

This scheme of assessing the student’s progress throughout the tutorial session distinguishes this tutorial system from other SQL tutors reported in the literature. An evaluation of students’ usage and acceptance of the tutoring system, within the Dublin City University campus, indicated that an impressive 20 percent of usage was during weekends. The paper is organized well, with figures providing much of the clarity. It is free from typographical errors.