I've just encountered the term "unselected sample" for the first time, which seems to be fairly commonly used in medicine. Example usages:
- Climacteric symptoms in an unselected sample of Swedish women.
- Post-partum mental disorder in an unselected sample. The psychiatric history.
- Test–retest reproducibility of accommodation measurements gathered in an unselected sample of UK primary school children
Does using the term "unselected sample" imply anything different from using the term "random sample"? After looking at a few sources, I'm a little confused:
Merriam Webster defines "unselected" as "chosen at random", and gives the phrase "unselected samples" as example usage. This would seem to imply that "unselected sample" is just a synonym of "random sample".
Psychology Dictionary says that "unselected sample" is "an unofficial terminology for a random sample."... but then seems to contradict itself by giving the example usage:
Unselected samples are often interchanged with random samples.
which to me seems to say that the two are subtly distinct concepts that are often wrongly conflated, not that they are literal synonyms.
The 1979 paper Representative Sampling, II: Scientific Literature, Excluding Statistics by William Kruskal and Frederick Mosteller quotes a confusing definition from 1919 given in terms of the "mode of variation" of the sample, whatever that means:
A sample "truly" represents a group when the mode of variation within the sample is the same as the variation within the group at large: this is what is meant when we say we have an "unselected" sample.
After reading these sources, I'm left unsure whether there's some difference between these terms - either an outright difference in definition, or perhaps some subtle difference in connotation that would lead an author to choose one over the other.