| Resumo: |
This work presents a new approach for automatic evaluation of SQL queries. This approach proposes a solution for a challenge to stimulate the student to improve his/her solution: searching, besides an answer that returns a correct result, a query with a complexity next to the optimum solution. This proposal can be used in e-learning environments or in presential education in laboratory activities, including the evaluations. The proposed solution has as advantages: (1) the student receives an instantaneous feedback during a practice programming activity, which allows the student to refactorize his/her solution in direction to a optimum solution; (2) complete integration between the programming concepts teaching with examples of fragments of on-line executable programs; (3) monitoring of students activities (how many examples were executed; in each exercise how many execution attempts were done, etc). This work is the first pass in direction to the construction of an environment, totally attended (for example with automatic evaluation) for teaching of SQL programming language, where the instructor is free of the hard work to correct SQL commands, being able to do more relevant pedagogic tasks. The method, based on statistics and Software Engineering metrics, can be adapted for other languages, such as Java and Pascal. In addition, the LabSQL serves as a laboratory for experimentation of two news techniques, one of evaluation and other of accompaniments that are being searched in several works at the same time: (a) automatic evaluation of discursive conceptual questions, besides to allow the traditional objective questions, (b) method of accompaniment through a montage of evaluation rubrics. |
|---|
