Chapter 5: Sampling and Generalizability

Test Bank

Multiple Choice

1. This subset of the population of interest is called a(n) ______.

A. element

B. sample

C. census

D. representation

Difficulty Level: Easy

2. Units that contain one or more elements and that are listed in a sampling frame are called ______.

A. elements

B. characteristics

C. enumeration units

D. sampling units

Difficulty Level: Easy

3. To assess sample quality when you are planning or evaluating a study, which of the following is NOT a question you need to ask?

A. From what population were the cases selected?

B. How large is the population?

C. What method was used to select cases from this population?

D. Do the cases that were studied represent, in the aggregate, the population from which they were selected?

Difficulty Level: Medium

4. What is another name for cross-population generalizability?

A. target population

B. sampling error

C. external validity

D. internal validity

Difficulty Level: Medium

5. The list from which elements of populations are selected for a sample is known as the ______.

A. population

B. sample

C. element

D. sampling frame

Difficulty Level: Easy

6. The entire set of individuals or other entities to which study findings will be generalized is the ______.

A. population

B. sample

C. target population

D. sampling frame

Difficulty Level: Medium

7. A subset of the population used to study the population as a whole is known as a/an ______.

A. target population

B. sampling frame

C. sample

D. element

Difficulty Level: Easy

8. How can a researcher test the validity of cross-population generalizations?

A. set higher confidence intervals

B. conduct more research in other sites

C. reduce sampling error

D. select a larger sample

Difficulty Level: Medium

9. In Metropolitan Region Y, 75% of the population lives in suburban areas, while only 25% remain in the central city. We draw a sample of 50 city residents and 50 suburban residents. This sample is ______.

A. invalid

B. unreliable

C. proportionate

D. unrepresentative

Difficulty Level: Hard

10. What new technique did the U.S. Census introduce in 2010 in an attempt to increase the response rate?

A. financial incentives

B. random sampling techniques

C. Internet-based response option

D. only face-to-face interviews

Difficulty Level: Easy

11. A survey that takes measurements from all elements of a population is also known as a ______.

A. representative sample

B. disproportionate representative sample

C. proportionate representative sample

D. census

Difficulty Level: Easy

12. Response rates in a random sample should be at least _______% in order to estimate a population parameter.

A. 30

B. 70

C. 50

D. 20

Difficulty Level: Medium

13. In the population of New Town, 30% of the people work for Bigg Corporation, 30% work in the public sector (including education), and 40% are self-employed. In a survey of 100 residents of New Town, 40 people worked for Bigg Corporation, 40% worked in the public sector, and 20 people were self-employed. This sample was ______.

A. overrepresented

B. underrepresented

C. nonprobabilistic

D. unrepresentative

Difficulty Level: Hard

14. If one third of all elements in a population will be selected for a simple random sample, the probability of selection for each element in the population is ______.

A. 0.05

B. 0.13

C. 0.33

D. 0.67

Difficulty Level: Hard

15. Which of the following is FALSE about probability sampling methods?

A. The probability of selection is known for all elements.

B. The probability of selection is greater than zero for all elements.

C. There is no systematic bias in selection.

D. There is no sampling error.

Difficulty Level: Medium

16. Which of the following is TRUE about probability sampling methods?

A. In most cases, the raw size of the sample is less important than the fraction of the population that the sample represents.

B. A smaller sample tends to be more representative of a population than a larger drawn from the same population.

C. A sample of 3% of the population is always more representative of the population than a sample of 2% of the population, regardless of the size of the sample drawn.

D. The more heterogeneous a population is, the more likely a sample will be representative of it.

Difficulty Level: Medium

17. When nothing but chance determines the elements selected for a sample, there is no ______.

A. sampling error

B. probability of selection

C. parameter

D. confidence interval

Difficulty Level: Medium

18. Which of the following is NOT a procedure for simple random selection of elements?

A. assigning consecutive numbers to elements and selecting on the basis of a random number table

B. selecting every nth element from a sampling frame, after the starting point has been randomly selected

C. using a computer to dial random digits after a phone prefix

D. writing all elements down on small pieces of paper, mixing them up, and selecting them like they do in the lottery

Difficulty Level: Hard

19. Which of the following is FALSE regarding simple random sampling?

A. Procedure identifies cases strictly on the basis of chance.

B. A random number table can be used to draw a simple random sample.

C. Computers can generate random numbers, cases, or phone numbers.

D. Simple random sampling must be done with replacement sampling.

Difficulty Level: Medium

20. A researcher gets a list of all 500 members of Social Club Z that she wants to include in her study. She only has the funding and time to survey 50 members. She takes her list of members, randomly selects a starting point, and then selects every 10th name from the list to be included in her sample. In this example, the sampling interval is ______.

A. the list of all 500 members

B. 500

C. 50

D. 10

Difficulty Level: Hard

21. Sampling bias known as periodicity occurs in which type of sampling design?

A. simple random

B. systematic random

C. stratified random

D. multistage cluster

Difficulty Level: Medium

22. A researcher studying small town religiosity in the United States randomly selected 10 states. From these states, he randomly selected one tenth of all counties. From these counties, he randomly selected one tenth of cities with populations smaller than 10,000. From these towns, he obtained lists of all houses of worship and randomly selected three. From these, he selected 10 practitioners to be interviewed. In this example, which is NOT a cluster?

A. states

B. counties

C. practitioners

D. houses of worship

Difficulty Level: Hard

23. Which two sampling methods require that the researcher know something about the salient characteristics of the population (such as race, ethnicity, or gender) before selecting samples?

A. simple random and stratified random

B. stratified random and quota

C. quota and multistage cluster

D. multistage cluster and purposive

Difficulty Level: Hard

24. When a 1936 Literary Digest poll found overwhelming support for Alfred Landon's presidential bid over Franklin Delano Roosevelt, they made an erroneous prediction that Landon would defeat FDR. What are the probable sources for error in their poll that helps explain their mistake?

A. Nonresponse rate was high.

B. Sample was not representative of the population.

C. Failed to use probability sampling techniques.

D. All of these

Difficulty Level: Hard

25. Which of the following is NOT a means for generating a random sample?

A. stopping the first people you meet on the street

B. flipping a coin

C. rolling a die

D. using a lottery method

Difficulty Level: Medium

26. Quota sampling is advisable when ______.

A. Salient characteristics about the population are known before the sample is drawn.

B. A sampling frame is unavailable.

C. Salient characteristics about the population are unknown.

D. Both A and B

Difficulty Level: Hard

27. A list of 1,000 students in alphabetical order has been provided to you, and you want a sample of 50. Using systematic random sampling techniques, what is the sampling interval?

A. 1/50

B. 1/20

C. 10

D. 20

Difficulty Level: Hard

28. In purposive sampling, a researcher should continue to select interviewees until which of the following things have been achieved?

A. Quotas have been met.

B. Completeness and saturation have been achieved.

C. Time and other resources have run out.

D. Probability of selection has been determined.

Difficulty Level: Medium

29. A researcher has decided to do a study of people who hunt in a nearby national forest. He asks a friend, whom he knows to be a hunter, if he would consent to an interview. He interviews his friend and then asks his friend if he could name other people who hunted in the forest. His friend provides him with five names. The researcher contacts those five people, interviews them, and asks each of those people for names of other hunters. This sampling technique is known as ______.

A. casual sampling

B. personal sampling

C. key informant interviewing

D. snowball sampling

Difficulty Level: Hard

30. You have a population of 1,000 students, of which 75% are male. You want to draw a sample of 100 students, of which 50 are male and 50 are female. What is the probability of selection for men?

A. 1/1,000

B. 75/1,000

C. 50/750

D. 100/750

Difficulty Level: Hard

31. Which of the following statements is FALSE regarding sample quality?

A. Sample quality cannot be evaluated if population is not specified.

B. Sample quality cannot be evaluated if selection technique is not specified.

C. Sample quality is determined by the selection method itself, not the actual sample obtained.

D. Cross-population generalizations are conjecture, no matter how strong the sample generalization.

Difficulty Level: Medium

32. Which of the following is FALSE regarding normal curves?

A. Sampling distributions for many statistics have a normal shape.

B. A normal distribution looks like a bell.

C. A normal distribution is asymmetric.

D. The shape of a normal distribution is produced by random sampling error.

Difficulty Level: Hard

33. Which of the following is NOT a convention for confidence intervals in the social sciences?

A. 80%

B. 95%

C. 99%

D. 99.9%

Difficulty Level: Hard

34. You have a population of 100, with 20% of them male and 80% female. The mean income for men is US$100,000 and for women is US$68,000. To estimate the mean income for the entire population, what is the weight you should use for the male population?

A. 10

B. 20

C. 40

D. 80

Difficulty Level: Hard

35. What is NOT another name for availability sampling?

A. purposive sampling

B. convenience sampling

C. haphazard sampling

D. accidental sampling

Difficulty Level: Medium

36. Rubin and Rubin (1995) suggest continuing to select interviewees until you can pass what test?

A. completeness

B. saturation

C. both A and B

D. none of these

Difficulty Level: Medium

37. Which of the following is NOT true about sample sizes?

A. The less sampling error desired, the larger the sample size must be.

B. Samples of more homogeneous populations can be smaller than can samples of more diverse populations.

C. When the researchers expect to find very strong relationships between the variables when they test hypotheses, they will need a smaller sample to detect these relationships than if they expect weaker relationships.

D. For professional studies of the national population in which only a simple description is desired, the minimum sample size required is 2,500.

Difficulty Level: Hard

38. Which sampling technique is useful for hard-to-reach or hard-to-identify populations?

A. availability sampling

B. snowball sampling

C. purposive sampling

D. quota sampling

Difficulty Level: Easy

39. Researchers can make more precise estimates of the sample size required through a method called ______.

A. statistical power analysis

B. generalization

C. inferential statistics

D. sample statistics

Difficulty Level: Medium

40. What is a method of sampling in which sample elements are returned to the sampling frame after being selected, so they may be sampled again?

A. simple random sampling

B. replacement sampling

C. systematic random sampling

D. convenience sampling

Difficulty Level: Medium

41. How is sample quality assessed?

A. the size of the sample

B. the amount of inference

C. the amount of sampling error

D. how thoroughly people in the sample answer the survey questions

Difficulty Level: Medium

42. A student researcher conducts a study examining binge drinking behaviors among college students. To do this, she conducts a survey with 1,500 college students from her university. She obtains survey respondents by stopping students in front of the library and asking them to complete her survey. What population can she generalize her findings to?

A. people who drink

B. college students

C. college students at her university

D. college students that took her survey

Difficulty Level: Hard

43. When should a researcher NOT use a sample?

A. When no list of the population is available.

B. When all units in the population are identical.

C. When the population is stratified.

D. When the population is not clustered.

Difficulty Level: Medium

44. Which of the following is NOT true about the generalizability of the Milgram experiments?

A. In replications, when another “subject” in the room who refused to administer the shocks, subjects were likely to resist authority.

B. Similar results to the original experiment were obtained in many replications of the Milgram experiments.

C. In the original experiment, all participants were men. Therefore, generalizing to women would not be appropriate.

D. The original experiment results were generalizable to the U.S. adult population.

Difficulty Level: Hard

45. Which of the following is NOT a reason that social scientists do not often conduct a census?

A. Conducting a census is too expensive.

B. Conducting a census is too time intensive.

C. A census can only be conducted by the U.S. government.

D. Using a sample is often good enough.

Difficulty Level: Hard

46. Which of the following is NOT one of the four most common methods for drawing random samples?

A. simple random sampling

B. systematic random sampling

C. cluster sampling

D. availability sampling

Difficulty Level: Medium

47. You should not assume that findings from a randomly selected sample will be generalizable to the population from which the sample was selected if the rate of nonresponse is above what threshold?

A.10%

B. 30%

C. 50%

D. 75%

Difficulty Level: Medium

48. As of 2013, what fraction of the U.S. population only cell phones?

A. 10%

B. 25%

C. 40%

D. 60%

Difficulty Level: Easy

49. Systematic random sampling is a variant of what sampling method?

A. simple random sampling

B. systematic random sampling

C. cluster sampling

D. availability sampling

Difficulty Level: Medium

True/False

1. A sample is always drawn from the target population.

Difficulty Level: Easy

2. The individual members of the population whose characteristics are to be measured are called the elements.

Difficulty Level: Easy

3. The difference between the characteristics of a sample and the characteristics of a population from which it was drawn is the sampling error.

Difficulty Level: Easy

4. The larger the sampling error, the more representative the sample.

Difficulty Level: Easy

5. Sampling is unnecessary if all units in the population are identical.

Difficulty Level: Medium

6. In a representative sample, some characteristics are overrepresented or underrepresented when compared to the population.

Difficulty Level: Medium

7. In a census, the probability of selection is ALWAYS less than 1.0.

Difficulty Level: Easy

8. Probability sampling methods have no systematic bias.

Difficulty Level: Easy

9. The fraction of the total population that a sample contains largely affects the sample's representativeness.

Difficulty Level: Medium

10. The validity of cross-population generalizations can be tested through sampling.

Difficulty Level: Medium

11. The U.S. Government uses tax dollars to conduct the Census.

Difficulty Level: Easy

12. Quota sampling requires that the researcher have some prior knowledge of characteristics in the population.

Difficulty Level: Easy

13. The proportion of the whole population that the sample represents is more important than the size of the sample for determining the sample's representativeness.

Difficulty Level: Medium

14. We can mathematically determine the likely degree of sampling error in an estimate based on a random sample, provided we have a reasonable response rate.

Difficulty Level: Easy

15. Qualitative research methods tend to use probability sampling methods.

Difficulty Level: Medium

Essay

1. What is the difference between cross-population and sample generalizability?

Difficulty Level: Hard

2. What is a normal curve? Why are normal curves important for inferential statistics?

Difficulty Level: Hard

3. Explain how both the size of the sample and the homogeneity (sameness) of the population affect the degree of error due to chance within a research study.

Difficulty Level: Hard

4. Identify and explain the three different types of random sampling.

Difficulty Level: Medium

5. What is snowball sampling? When would this type of sampling method be appropriate to use?

Difficulty Level: Hard

6. What is the difference between probability and nonprobability sampling?

Difficulty Level: Hard

7. In social research, what is the importance of sampling? Why are probability sampling techniques superior to nonprobability sampling in achieving this end?

Difficulty Level: Medium

8. Describe how you would select a sample of students from your college, using random sampling techniques.

Difficulty Level: Hard

9. Identify and explain the four types of nonprobability sampling methods. Provide examples of each. Why are these techniques inferior to probability sampling techniques?

Difficulty Level: Hard