Social vulnerability and COVID-19 in Maringá, Brazil

Published: 17 December 2021| Version 1 | DOI: 10.17632/kcfnsyt72f.1
Eduardo Alcantara Ribeiro, Udelysses Janete Veltrini Fonzar, Ícaro da Costa Francisco,
, Matheus Pereira Libório


The available data were used to investigate the relationship between social indicators to cases and deaths from COVID-19, since this relationship is sensitive to geographic scale and failures due to inadequate indicators and outdated population data.


Steps to reproduce

Two methods were used to construct the composite indicators of social vulnerability: Benefit-of-the-doubt and Principal Component Analysis. COVID records were georeferenced and superimposed on the polygons of the census tracts. Then, the census tract code was associated with each COVID record. This association was carried out in QGIS through the union between vector layers. The census tract code is then used to link the composite indicator score for social vulnerability to the COVID record. The data is organized as follows: 1.) Social vulnerability: data of fifteen social indicators for each um two 551 urban census tracts of Maringá, Brazil. The data relative to the size of the population and the number of household units. The raw data can be obtained without Related link 1. 2.) Composite indicators: data referring to the composite indicators built using the Benefit-of-the-doubt and Principal component analysis methods using the fifteen social indicators listed in the "Social vulnerability" spreadsheet. 3.) Covid: COVID-19 records from March 18, 2020 to March 17, 2021. These data refer to the number of cases and deaths from COVID-19 with no identification of people by census tract.


Pontificia Universidade Catolica de Minas Gerais, Editora da Universidade Estadual de Maringa


Geography, Statistics, Health