ADoTe: abordagem para o ensino e aprendizagem de critérios de teste da técnica funcional apoiada por dojo de teste

Abstract

Context: Despite the importance of software testing and its presence in academic curricula, students of computer science and related courses (including those who have already graduated) often have knowledge gaps on testing and low levels of motivation and interest in learning this subject. Functional testing technique criteria, for example, are usually required and used in industry, but their teaching still needs support. Additionally, there has historically been a shortage of qualified testing professionals. Faced with the problems caused by such gaps, improvements in testing education are demanded by the industry for the academy. In this scenario, alternative approaches (such as active, collaborative, and cooperative learning strategies) have been the subject of study, being dojos adherent to such strategies. However, although dojos have the potential to contribute to solving the aforementioned problems (since positive results are usually reported regarding their impact on students’ learning and motivation), their use is still little explored in the software testing area. Objective: In this context, this study aimed to define and evaluate an approach supported by testing dojo for theoretical and practical teaching and learning of functional testing technique criteria in higher education. Methods: The approach (named AdoTe) was defined iteratively, supported by information from the literature (including a systematic mapping (SLM) regarding the use of dojos in computing) and by quali-quantitative analysis of the results of two operations of a controlled experiment that evaluated the impact of AdoTe (compared to a traditional teaching approach) on students’ learning (measured by knowledge tests) and motivation (measured by the instrument known as IMI). Results: From the analysis of data collected from 44 participants it was evidenced that, although the averages for learning and general motivation of the groups submitted to AdoTe were higher than those of the groups submitted to a traditional teaching approach, it was not possible to state that there is a statistically significant difference between the approaches evaluated (for these variables). However, the results of the IMI subscale related to intrinsic motivation were statistically significant, indicating that students feel more interest/enjoyment in learning functional testing technique criteria through AdoTe (the interest/enjoyment averages of the groups submitted to AdoTe were 15.80% and 19.71% higher than those in the control groups). Additionally, the results of thematic analyses carried out on the responses to a retrospective questionnaire showed that AdoTe was well accepted by students and reinforced the importance of its steps and guidelines. Conclusion: The approach and the SLM presented in this work contribute to the set of studies that aim to improve software testing education. Furthermore, AdoTe positively impacts students’ learning and motivation and, compared to traditional teaching, tends to lead to greater levels of interest/enjoyment in learning functional testing technique criteria.