ABSTRACT
The two competing hypotheses in a hypothesis test are called the null and the alternative. It is important to be able to tell which is which, because they are not simply mirror images of each other. The hypothesis for which you are seeking proof is the alternative. However, the company would be concerned about being off either way—it would be just as concerned about overweighting as underweighting, and it would not want to commit a type I error in either direction. Although hypothesis testing is generally presented as a separate topic from estimation, the two inferential procedures are really applications of the same basic reasoning. So the company would worry about the possibility of getting a statistic in either tail of the distribution, and the p value would be the sum of the proportions in both tails.
