Hey! Just saw this post and your comment. Most research in stuttering uses VERY small samples. We think this is unfortunate because we probably aren't getting a very good picture of the population of people who stutter, though for some research small samples are ok (it depends on the question and the study). But, the biggest surveys in stuttering are only a couple hundred responses and most of them are skewed to people who have had therapy or participate in self-help or support. There's so much we need to learn about the population of people who stutter. It's particularly hard finding people who are unconnected--people who haven't had therapy, or aren't connected to other people in other ways (like this reddit).