Data samples in which you expect no change to occur

The answer is negative controls. These are groups where no phenomenon is probable. They ensure that there is no effect when there should be no effect. Where there are only two possible outcomes, e.g. positive or negative, if the treatment group and the negative control both produce a negative result, it can be concluded that the treatment had no effect. If the negative control group and the treatment group both yield a positive result, it can be inferred that a puzzling variable is involved in the occurrence under study, and the positive results are not only due to the treatment. In other examples, outcomes might be measured as lengths, times, percentages, and so forth.