Research methods

A probability sample estimate differs slightly from the true population value simply because a subset rather than the whole population was observed. What is this difference called?