You wrote, "To have any kind of conclusions to draw from this study, many more Johnstons would have to take it, then compare results."
That's not exactly true but it does depend on what conclusion you want to draw. One thing the study has proved over and over is that not all people with the surname are related. In some of the people tested it is pretty certain they have not had the same male ancestor for thousands of years.
DNA is very good at proving negetive relationships.
DNA is frequently used to prove that a certain person did not commit a crime. That is on the news every night.
However, a match between two DNA samples is not absolute proof that the samples are from the same person. That means it is stated as a probability, not a certainty. Anybody who does a paternity test gets a probability of paternity, like 1 in 13,000,000 or something.
Genealogy testing is not providing proof of a relationship. It is just another piece of evidence.
When you find a name you are looking for on a census that is not proof you have found the person you are looking for. It could be another person with the same name. However if it is the right county and the wife and kids names and ages match what you expect you can be pretty confident you have found the right person. Unfortunately you have not proved it. That's the way it is with DNA. It's just another fact you look at.