Análisis de Correspondencias Múltiples en donantes de sangre de la Cruz Roja Ecuatoriana Junta Provincial de Chimborazo, periodo 2017-2020

Abstract

The objective of this work was to study the interdependence between demographic variables in blood donors at Cruz Roja Ecuatoriana Junta Provincial de Chimborazo during 2017-2020, through the application of statistical techniques. The information matrix was taken from the e-Delphyn blood donor registry software, where the information collected in blood donation campaigns is stored. For the study, the considered variables were: sex, age, population, new donor, type of donation, blood group, acceptance of the donation. A descriptive analysis was carried out on each variable under study to observe the behavior of the data, then a correlation analysis was applied to verify the existing relationship between pairs of variables. After that an independence analysis was performed to observe whether or not there is a relationship between the variables under study. Finally, a multiple correspondence analysis was applied in order to represent in a reduced multidimensional space the relationship between the categories of the qualitative variables proposed in the present study. The results were that only the variables age and new donor presented a relatively weak linear relationship, inversely proportional. The chi-square hypothesis test of independence identified that the pairs of variables type of donation and sex, type of donation and population, blood acceptance and population, were independent of each other; The application of multiple correspondence analysis through the perceptual map did not generate optimal results for the study, since the percentage of variability of the first two main coordinates was 10.71%. It is concluded that the variables sex, population, age and new donor contributed significantly to blood donation. It is recommended to apply multivariate techniques in future research to contribute to decision making.