The mean of the sampling distribution is 0 since the assumption is that there are no differences between population means and therefore the average difference between sample means would be zero. The graph of the sampling distribution shows that a difference between sample means of two or more is not an improbable occurrence even when there is no difference between population means. Twenty percent of the time the Drug group would be two or more points lower and 20% of the time the Control group would be two or more points higher. How should the results of the experiment be interpreted in light of this analysis? Science requires a conservative approach to decision making: a conclusion is accepted by the scientific community only if the data supporting it are strong enough to convince a skeptic. No skeptic would be convinced that the antidepressant drug rather than chance caused the difference.