Suppose that you are asked to improve the accuracy of the U.S. Census. The census forms are mailed to all households. Suppose that only about half of all households return the forms. To improve accuracy should you (a) have census-takers visit the other houses up to three times, hoping that people are home and will fill out the forms? or (b) take a random sample of the households that did not return forms and visit those households as much as necessary to get the forms filled out from that sample? Justify your answer.
The census-takers believe that having more children causes households to own more televisions. What is the explanatory variable, and what is the response variable? Is this an experiment or an observational study? What are some lurking variables? Are any lurking variables likely to be confounded with the explanatory variable?