Guia para decompor tarefas de testes funcionais em software crowdsourcing

• Ano de defesa: 2020
• Autor(a) principal: Costella, Fernando Luiz
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade de Passo Fundo; Instituto de Ciências Exatas e Geociências – ICEG; Brasil; UPF; Programa de Pós-Graduação em Computação Aplicada
• Palavras-chave em Português: Engenharia de software; Software de aplicação; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.upf.br/handle/123456789/1809

Participants in crowdsourcing environments face problems, which are considered barriers to make their contribution effective. One of the difficulties presented is the complexity of the tasks provided. The work presents an approach to reduce this complexity, by decomposing tasks. Unlike related works that demonstrate the decomposition by the platform, the proposal foresees that it is the user who performs this procedure. For this, a guide for decomposing Macrotasks was developed, incorporating in its elaboration the patterns identified in the tasks by content analysis, the Hierarchical Task Analysis technique and the Test Process Improvement model. For the evaluation of the guide, a paired experiment was carried out, with students selected for convenience of the Software Testing discipline at the University of Passo Fundo. The experiment was divided into two stages: in the first, software testing tasks were performed without using the guide and in the second, with different tasks from Phase 1, using the proposed guide. The experiment showed insignificant results in the quantitative analysis and significant results in the qualitative analysis. It is concluded that even though the quantitative analysis did not present relevant results, the qualitative analysis demonstrated a good adherence of the participants, demonstrating the importance and relevance of the guide.