Ch 4 extra Review

What is the purpose of the random assignment of treatments to subjects in an experiment?

• To create roughly equivalent groups before treatments are administered.

• To prevent certain groups in the population from being systematically excluded from the experiment.

• To reduce bias in the selection of the subjects to be included in the experiment.

• To eliminate the need for blinding in an experiment.

A random number generator is used to select 12 students from a large statistics class to rate a statistics video. The 12 students selected are

• the population.

• a simple random sample of the class.

• a convenience sample.

• a voluntary response sample.

A survey consisted of a random sample of 3800 U.S. adults. The adults were asked, "Should we have a third major political party in this country in addition to the Democratic and Republican parties?" In this sample, 53% responded, “Yes, we should have a third major political party.” What is the largest population to which these results can be generalized?

• all registered voters

• all adult residents of the U.S.

• the 3,800 people who were interviewed

• all people who think we should have a third political party

A study showed that teenagers who had night lights in their rooms as babies are more likely to suffer from myopia (near sightedness) than teenagers who had not used night lights. But genetic factors are known to play a role, and near-sighted parents are more likely to install a night light in their children’s room. This makes it hard to tell if the night light or genetics caused the myopia. This is an example of

• a response variable.

• the placebo effect.

• confounding

• bias

Which of the following is a valid statistical question?

• How many people voted in the last school board election?

• How many points were scored by the New England Patriots during the 2019 NFL season?

• How long do students typically wait in line when getting lunch from the cafeteria?

• How tall is Mount Everest?

The most important advantage of experiments over observational studies is that

• experiments are easier to carry out.

• experiments can give better evidence of causation.

• an observational study cannot be generalized to a larger population of interest.

• an observational study cannot have a response variable.

A local community organizer is considering running for city council. She wants to measure her name recognition in her community by conducting a survey in the district in which she intends to run. Which of the following survey methods would produce unbiased results?

• Place pollsters in front of every grocery store in her district on Saturday morning and give them flyers with her name and picture on it. They will show the flyer to each person entering the store and ask if they recognize her name.

• Send a survey card to all registered voters in her district asking them to call a number to state whether or not they recognize her name.

• Call people from across the country using random digit dialing and ask each person who answers if they recognize her name.

• Select a random sample of eligible voters in her district, have a pollster visit each of their homes and ask if they recognize her name.

The organizer of a 5K race wants to know how many miles per week runners typically run. To investigate, she surveys the first 30 people who check-in at her race. Which of the following statements is true?

• This is a voluntary response sample. It will likely overestimate the number of miles that runners typically run.

• This is a voluntary response sample. It will likely underestimate the number of miles that runners typically run.

• This is a convenience sample. It will likely overestimate the number of miles that runners typically run.

• This is a convenience sample. It will likely underestimate the number of miles that runners typically run.

In the late 1990’s, Scotland was considering independence from England. One random sample showed that 51% of Scots favored "independence." Another random sample selected at the same time showed that only 34% favored being "separated" from England. Why are these results so different?

• These results most likely differ just by chance due to random sampling.

• These results most likely differ because of the difference in the wording that was used.

• These results most likely differ due to nonresponse bias.

• These results most likely differ due to undercoverage bias.

What is the main reason for using a control group when designing an experiment?

• to provide a baseline for comparing the effect of the treatment(s)

• to generalize the results to a larger population

• to reduce sampling variability

• to create roughly equivalent groups at the beginning of the experiment