In the case of ordered samples, every permutation of a sequence is considered unique. This is applicable when the order matters. The formula given for this is , where n is the number of options for each choice, and k is the number of choices.
For unordered samples, it is not the sequence but the combination of choices that matters, and permutations of the same combination are considered equivalent. The naive counting would use , but this can lead to overcounting as not all permutations are distinct in this context. The correct formula for unordered samples is , which is derived from combinatorial considerations that account for the indistinguishability among permutations.
The text mentions Bose-Einstein statistics, which apply to quantum particles known as bosons, where particles are indistinguishable and multiple particles can occupy the same state. However, the text argues that this approach is not suitable for calculating probabilities like the birthday problem, where the items (birthdays) are distinguishable and the ordering does not matter.