Question

Reporting weight loss A total of 300 people participated in a free 12-week weight-loss

course at a community health clinic. After one year, administrators emailed each of the

300 participants to see how much weight they had lost since the end of the course. Only

56 participants responded to the survey. The mean weight loss for this sample was 13.6

pounds. Describe how nonresponse might lead to bias in this study. Explain the likely

direction of the bias.

Short answer

Respondents to the survey are more likely to be those who dropped the greatest weight (since they are pleased of their accomplishment and want to share it), implying that our estimate of the mean weight loss is likely to be too high, and the real mean weight reduction is most likely less than $13.6$pounds.

Step-by-step solution

:given information

We need to describe how nonresponse might lead to bias in this study.

Explanation

The weight reduction of the $300$participants is of interest to us, and $56$people answered to the survey.

Nonresponse bias is present in our sample since it excludes those who did not complete the survey.

Respondents to the survey are more likely to be those who dropped the greatest weight (since they are pleased of their accomplishment and want to share it), implying that our estimate of the mean weight loss is likely to be too high, and the real mean weight reduction is most likely less than $13.6$pounds.