Resumo

SAGARO DEL CAMPO, Nelsa María  e  ZAMORA MATAMOROS, Larisa. Implicative statistical analysis versus binary logistic regression for the study of causation in health. Multimed [online]. 2019, vol.23, n.6, pp. 1416-1440. ISSN 1028-4818.

The purpose of this paper is to establish a comparison of two multivariate statistical techniques used in clinical-epidemiological research to identify prognostic or risk factors from observational designs. Binary logistic regression, widely used in health since the middle of the last century, is compared to identify the influence of various factors on a dichotomous outcome and the implicit statistical analysis, a data mining tool, used to model the quasi-implication between events. and variables, which arose to solve problems of the Didactics of mathematics; for which a review of the literature and of the investigations in which both techniques were applied simultaneously was carried out. Fourteen comparison patterns were defined. The advantages of the implicative statistical analysis are presented and its contextualized use is suggested prior to the logistic regression in the epidemiological studies of causality.

Palavras-chave : Logistic regression; Implicative statistical analysis; Quasi-implication; Similarity; Cohesion.

        · resumo em Português | Espanhol     · texto em Espanhol     · Espanhol ( pdf )