Correct Response: D. Best practice for a well-designed experiment is to include a test group for which the factor being tested (i.e., the independent variable) is not applied. In all other aspects, the conditions for the control group are identical to the groups to which the independent variable is applied. The results from the control group provide baseline data that allow researchers to determine whether the application of a single independent variable has an effect. Ensuring the reproducibility of the results (A) is best accomplished by keeping detailed records of the set-up and of the actual conditions of the experiment. This allows for the experiment to be faithfully replicated in order to determine whether the results are reproducible and real rather than a statistical anomaly. Although peer reviewers will most likely look for the inclusion of a control group when they evaluate the experiment, inclusion of a control group does not facilitate the peer review process (B). Potential bias may be reduced (C) by practices such as double-blind experiments in which both the participants and the observers are unaware whether they received the actual treatment or a placebo. However, reduction of potential bias is not the primary reason for the inclusion of a control group.
