Uso de Learning Analytics sobre dados de questionários para apoio à avaliação da aprendizagem por professores e gestores

Abstract

Assessment is an important step in the teaching and learning process, as it is possible to measure students learning about the content covered. The evaluation of students through the application of multiple-choice questionnaires in virtual environments is one of the teacher’s tasks, however, the learning evaluation process is restricted to information that is commonly provided by the platform, and in general this information is restricted to score obtained by students according to the number of questions correctly marked. The large amount of data generated by students from the use of these systems can provide a much larger set of useful information for teachers and managers, however, searching and analyzing that information becomes an unviable task to be performed manually, in view of the volume of data available. In this regard, this project used Learning Analytics techniques to generate analysis and visual structures that facilitate the understanding of teachers and managers about student interaction data when answering multiple choice questionnaires in a virtual environment, using different metrics for assessing learning as a basis to go beyond the traditional note. For automation and presentation of analyzes and visualizations, a module was developed that uses Learning Analytics and provides students learning information by class, discipline, topics and question. The evaluation of the tool by the teachers and pedagogue brought positive results on its usefulness and intention to use, although there is some deficiency in relation to the interface.