Estimation of the correlation between a continuous and a discrete variable

Abstract

The maximum likelihood (ML) procedure is used to estimate the correlation between a discrete and a continuous variable. The discrete variable is assumed to be a classification of an unobservable continuous variable whose joint distribution with the observed continuous variable is bivariate normal. As an illustration, the correlation between the bodyweight at the time of mating, and the number of lambs born in a flock of ewes is estimated.