- This is the approach to probability adopted by classical statisticians who estimate the probability of an event (for example, the probability that a given individual x will have disease y) by taking a suitable sample of the relevant population, discovering the prevalence of the disease in that sample, and then inferring that the chances of x having y are the same as the prevalence rate in the sample. Strictly speaking, the frequentist approach depends upon a 'large' number of samples being taken. This approach is to be contrasted with the Bayesian approach. Much heat has been generated as to which approach is more useful (there is agreement about the maths) - as may be expected, given that each approach depends upon particular (subjective) assumptions being made and holding true, though the nature of these assumptions is not the same in each. For example, the frequentist approach involves judgments about how many samples to take, and of what size, and upon a judgment that the sample is a sufficiently faithful representation of the population.