A Set of things, whose collective properties we are interested in studying in Statistics. One can model properties of this set (like Frequency distributions) as probability distributions, even though these don't have the interpretation of probabilities, as there isn't any randomness. A more rigorous analysis, usually combines a population with a random Sampling procedure, which defines Random variables known as Samples. Functions of these samples are known as Statistics, which gives statistics its name.

–> Sometimes the population itself is modelled as a Random variable too. This is because one can interpret it as having uncertainty too, particularly in the Bayesian statistics approach. In that case the concept of population and sample mix a bit; or perhaps one is just calling "population" what we would typically "sample" (i.e. outcome of a random experiment); at the end there are fuzzy areas, and one may find "population" being applied to describe any random variable ..

In Statistical physics, there is an equivalent concept, applied to physical systems, known as Statistical ensemble