Answers to Test Yourself

Below are answers to the chapter Test Yourself feature.

(1) a) Since a simple random sample would be too difficult to take, I would try to take a cluster sample. It would be important to get people from across the country to weigh in. That is, people in urban and rural settings, or maybe more to the point, people living in areas with a high number of immigrants and people living in areas with a low number of immigrants. b) You could conduct a simple random sample of seniors at your university or college or you could do this for a number of universities/colleges of different types c) Again, a simple random sample of “people” is too broad, even if it’s narrowed to adults in this country. A type of convenience sample, like a volunteer sample, might not be a bad idea. You’re probably looking for a lot of people and not too worried about demographics. (2) b (3) b (4) e (5) a (6) a (7) b

(8)  Probability samples are typically better representatives of the population they are drawn from because the probability of an individual being chosen is determined ahead of time and controlled through random sampling. (9) Many studies use convenience samples because it can be difficult to identify all the members of large populations to randomly select them. (10) There are several ways to do this, but the key is to use a convenience sample of students at the school in the same proportions as they exist at the school as a population based on years in school and gender. These proportions will need to be known before the sample is selected. You will also need a recruitment technique (Internet with e-mail contact, surveys mailed to campus address, etc.) to describe how to administer your survey. (11) Internet samples tend to be more diverse than other samples; they can be collected using fewer resources and in a shorter period of time. (12) d, (13) b, (14) b  (15) population, sample (16) simple random