Chapter 23

23.1. We would like to know both the sample size and the actual mean weight loss before deciding whether we find the results convincing. Better yet, we would like to know exactly how the study was conducted and to have the actual data. Unfortunately, in many research studies, it is not possible to get the actual data from researchers.

23.2. No. If all 122 null hypotheses of no difference are true, we would expect 1% of the 122 (about one) of these null hypotheses to be significant at the 1% level by chance. Because this is consistent with what was observed, it is not clear if chance explains the results of this study.