In a typical two stage procedure, breeding value prediction for calvin
g ease in a threshold model is conditioned on estimated genetic and re
sidual covariance matrices. These covariance matrices are traditionall
y estimated using analytical approximations. A Gibbs sampler for makin
g full Bayesian inferences about fixed effects, breeding values, thres
holds and genetic and residual covariance matrices to analyze jointly
a discrete trait with multiple ordered categories (calving ease scores
) and a continuously Gaussian distributed trait (birth weights) is des
cribed. The Gibbs sampler is implemented by drawing from a set of dens
ities - (truncated) normal, uniform and inverted Wishart - making impl
ementation of Gibbs sampling straightforward. The method should be use
ful for estimating genetic parameters based on features of their margi
nal posterior densities taking into full account uncertainties in esti
mating other parameters. For routine, large-scale estimation of locati
on parameters (breeding values), Gibbs sampling is impractical. The jo
int posterior mode given the posterior mean estimates of thresholds an
d dispersion parameters is suggested. An analysis of simulated calving
ease scores and birth weights is described.