Ecological Systems and Devices Annotation << Back DEFINING STATISTICAL RELATIONSHIPS BETWEEN QUANTITATIVE AND DICHOTOMOUS VARIABLES: THE LOGISTIC REGRESSION AND MOVING AVERAGE METHODS A.N. Varaksin, Yu.V. Shalaumova, T.A. Maslakova, E.D. Konstantinova A new method for assessing the statistical relationship between the quantitative variable Y and the dichotomous variable X, which takes two values: X = 0 and X = 1, is proposed. The technique is based on dividing the quantitative variable Y into strata using the moving average method and calculating the average values in strata for the variables Y and X. As a result of stratification, the dichotomous variable X turns into a quantitative variable. After such a transformation of the variable X, the statistical relationship between Y and X can be analyzed by linear regression methods. Thus, the proposed methodology expands the set of methods for analyzing statistical relationships between quantitative and dichotomous variables. The use of moving average methods also makes it possible to visualize the relationship between the studied variables, which is important for applied applications, for example, in medical and environmental research. This paper compares the new methodology with one of the main existing methods for studying statistical relationships between quantitative and dichotomous variables: the logistic regression method. It is shown that the moving average method allows not only to estimate the statistical relationship between Y and X, but also to check the applicability condition of logistic regression. Keywords: quantitative and dichotomous variables; logistic regression; moving average method; linear regression; analysis of variance; conditions of applicability of logistic regression. DOI: 10.25791/esip.04.2021.1219 Pp. 03-09.