Author Question: An ABC news poll over the Internet resulted in 17,251 responses. Explain why the results of this ... (Read 779 times)

Shelles · **on:** Jul 24, 2018

An ABC news poll over the Internet resulted in 17,251 responses. Explain why the results of this poll should not be generalized to the American population.

Question 2

Explain why a cause and effect relationship can be inferred from a randomized experiment, but not from an observational study.

ngr69 · **Reply #1 on:** Jul 24, 2018

Answer to Question 1

The 17,251 responses in the sample were only made by people who have access to the Internet: not every person in the American population has internet access. The sample was not representative of the population; therefore the results from the sample should not be generalized to the population.

Answer to Question 2

In a randomized experiment, the effects of confounding variables should be similar in each group: the only systematic difference between the groups is the treatment (explanatory variable). Any observed differences in the response variable should result only from the difference in treatments. In an observational study, the groups defining the explanatory variable occur naturally and they may have other natural differences between them in addition to the explanatory variable. A confounding variable is always a possible alternative explanation for differences seen in an observational study.