If we test subjects receiving treatment A in one room and subjects receiving treatment B in another, we have created an orderly, or systematic, change in many of the variables that make up the testing conditions. We will not be able to tell for sure whether the independent variable or something about the different testing rooms produced changes in the groups. The A group may do better if it is tested in the same sunny room as the B group is. The key to controlling variables that cannot be eliminated or held constant is the third technique for physical variables balancing.