h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

It's worth adding that the concept of the control groups is very important in designing experiments particularly in the biomedical field. For example, in testing a new drug on a group of patients, for any results to be valid there should be a matched group of patients who do not get the drug (or usually get a placebo). Otherwise it's always possible to argue that an observed effect was not due to the drug and it would have been seen anyway even in untreated patients. This would have applied to your "violent videos" experiment where a control group of children would have had to been shown to have never been exposed to this material to act as a comparator group (making the whole experiment even more difficult).

Hiya 210967 (you really ought to change that to something easier )

Control groups are indeed very important in medical and psychological research. The entry Things to Consider when Reading Medical Research (A852761) covers this and many other things. But I don't think this entry needs to mention them, and it's already on the front page, so it would have to wait for the slow mill of the update process even if it was needed.
Control groups are one of many possible elements in well formed experiments in different disciplines that weren't mentioned. I skipped past it by saying that "One requirement is that the predictions must support the hypothesis" and pointing out that this was not always easy to determine. Medical research and control groups could have been a good example.

Hi, can't read right now, marking it for later