Analysis of Bad Smells in Programming with Dr. Scratch

Abstract

This project consists of the evaluation and analysis of the impact which bad habits have in the programming area. In particular, in this work we want to analyze the effect of “bad smells” in the development of the Computational Thinking skills. For its development, we have based on the Dr. Scratch tool, a free software web application which allows the analysis of projects designed with Scratch -programming language oriented to education- and to obtain an assessment about different aspects related to the Computational Thinking. The final objective of the project is the implementation of a new assessment model in Dr. Scratch which allows to raise awareness and prevent about the use of “bad smells” in programming with Scratch. In order to carry out this process, several phases of work have been necessary, with different technologies involved. In an initial phase, it was needed an update of the Dr. Scratch tool. The technologies used for that were related to web programming and the cloud production environment, such as Django, MySQL, Microsoft Azure or Google Cloud Platform, among others. During the second phase, we carried out an exhaustive analysis about “bad smells”. For this procedure, we used Jupyter Notebook, an appropriate technology for the data analysis. For the web design of the new model described previously, which constitutes the third phase, technologies such as HTML, CSS and Bootstrap were used. Finally, in order to verify the effectiveness of the project, we designed and implemented an assessment experiment with different teachers. To achieve this last phase, Google Forms together with the Dr. Scratch tool were used.