Lyzeli: a tool for identifying the clues in survey research data

  • João Pedro Rodrigues UFBA
  • Nabor Mendonça UNIFOR
  • Ivan Machado UFBA

Resumo


Survey research is one of the most used empirical research methods, and it is a means to contribute to theory development. A survey can be exploratory, descriptive, or explanatory. Whatever the objective, it will generate data that may support the construction of a theory or validate it. However, researchers must carefully evaluate data so they can transform them into useful information for science. Little is said about how complex and susceptible to human failures this process can be. We note that all the data analysis work is still highly manual and exhausting for the researcher, who is not always qualified to perform data analysis and, sometimes, can make mistakes in this process, inserting wrong numbers or calculating something wrong. Some existing tools can support this process but do not specifically focus on analyzing survey responses, combining valuable qualitative and quantitative analysis of responses. In this study, we introduce Lyzeli, a tool designed to assist researchers in analyzing and correlating survey responses. The proposed tool aims to deliver automatic detection of question types, sentiment analysis of responses, data filtering, word cloud, word count, graphics, codification for open-based questions answers. Tool Demonstration: https://youtu.be/e2w4TfhlxBY

Publicado
29/09/2021
RODRIGUES, João Pedro; MENDONÇA, Nabor; MACHADO, Ivan. Lyzeli: a tool for identifying the clues in survey research data. In: SIMPÓSIO BRASILEIRO DE ENGENHARIA DE SOFTWARE (SBES), 35. , 2021, Joinville. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 .