In Bayesian analysis, if there is no a priori evidence to suggest that one member of the population is any different from other members with respect to the property to be measured, the members are said to be exchangeable in the sample.
0