Ferramentas estatísticas na avaliação de serviços de abastecimento de água em áreas rurais

Abstract

One of the agendas for current discussions is the lack of basic sanitation in rural areas. This study aims to scrutinize twenty bibliographic references (articles, dissertations and theses) from different countries, between the years 2006 to 2022, which address water supply in rural areas as their main line of research. As a method of analysis of relative efficiency, the Data Envelopment Analysis (DEA) was used. A non-parametric methodology was chosen. Inputs and outputs were listed, divided into three groups: G1, G2 e G3. The water supply in groups G1 and G2 showed 100% efficiency, while group G3 showed 22% efficiency. For descriptive statistical analysis of the data, the Jamovi software, version 2.3.21.0, was used. The Shapiro-Wilk p results obtained were less than 0.05, which demonstrated that the data do not have a normal distribution. After statistical data analysis, Friedman's non-parametric test was chosen, in which the calculated p-value was above 0.05, indicating that there are no statistically significant differences when comparing the three groups. The DurbinConover test was used as post-hoc and there was no significant difference between them. Examining them, it could be seen that the G2 group showed better results than the G1 and G3 groups, representing better efficiency in communities that pay a monthly fee for system maintenance; that management is done by the community itself and that there is effective water treatment. In this way, a satisfactory water supply system/solution, after the results obtained from this study, is understood to be one that is managed by the community with government support, that the community pays monthly fees to cover maintenance and that there is water treatment and government technical guidelines.