Chapter 4.1, Problem 28E

To determine

To describe how non-response might lead to bias in this study and explain the likely direction of the bias.

Explanation of Solution

A survey of drivers began by randomly sampling from all listed residential telephone numbers in the United states. Of $45956$ calls to these numbers, only $5029$ were completed. They wanted to survey how far people drive on average per day.

Since the data surveyed due to less number of responses were very small.

i.e., $\begin{array}{l}\frac{5029}{45956}=0.1094\\ =10.94\%\end{array}$

thus, only $10.94\%$ of the total number of calls made responded to the survey. And this is less than $0.5$ which means that the sample responded cannot correctly defines the result of the survey. It does not represent the true population that is to be surveyed. Thus the non-response of the people lead to the biased result of the survey and we cannot conclude according to the response we get from it.