What you would get is a pretty good picture of how things work in the real world and not within the carefully controlled confines of a lab or clinical setting. Of course, it also depends on what you are trying to study. If the goal of the study was to see which diet was easier to stick to, they would have been better off allowing people to pick the plan that most appealed to them since sticking with something you find distasteful is not going to work very well. But...since (I believe), the goal of the study was to see which diet was more effective (and being able to stick with the plan is a big part of that), I can see where they might have felt it better to randomly assign dieters even if it doesn't match "real life" experience. I just happen to think it was flawed since being able to stick with your plan plays a big part in how effective it will be for you.
In the real world, people pick the diet that they want to try. In the real world, they get as much (or little) information as they are willing to look or ask for.
Since there is an addictive/emotional component to sweet/carby foods, giving someone a diet sheet and telling them to go forth and lose weight is about as effective as giving someone an instruction sheet on alcoholism and telling them to go get dry. As we see daily here on the forum, support and accurate information are vital to success. Quite honestly, I'm a bit surprised that the dropout rate wasn't higher in these studies on either side; on the low carb side due to lack of support and/or information and on the low fat side just due to sheer hunger or on either side from being matched with a plan that was unappealing to the dieter.