@umbra21 - I think the idea of a control group is a valuable one in almost any case. It is a relatively fundamental concept to have a control group vs an experimental group even if it doesn't always look exactly like a traditional science experiment.
But there are definitely famous experiments which didn't use this kind of control and they really should have.
The one that springs to mind is the Standford Prison experiment, where they took groups of men and arbitrarily assigned them as prisoners and guards and watched how they changed psychologically. They always say it proves something about the nature of power, but it was a very limited experiment. The men were similar in age and background, not to mention the same gender, and there was no group in similar conditions without the same power structure to compare what would happen. Without that kind of control group, we don't know how much was due to the power imbalance and how much was due to other factors.