Deck 11: Survey Studies

Question 1

The precision with which a sample approximates its population can be estimated using the inverse of the:

A) sampling frame
B) sampling error
C) sampling parameter
D) sampling fraction

Answer 1

B

Question 2

When conducting a study, the sampling frame refers to:

A) the list of all people selected to participate in a study
B) the list of all members in a population of interest
C) the list of only those people who were randomly chosen to be contacted for participation
D) the list of people who volunteered to take part in a study

Answer 2

B

Question 3

The population parameter in a study is:

A) the standard deviation for the population
B) the list of a population from which researchers draw their samples
C) the accuracy with which a sample approximates its population
D) the population's "score" that a sample is meant to approximate

Answer 3

D

Question 4

_____(a)_____ is of primary importance in laboratory experiments, _____(b)_____ is of primary importance in survey studies.

A) random assignment; random assignment
B) random assignment; random sampling
C) random sampling; random assignment
D) random sampling; random sampling

Question 5

Which of the following sampling techniques would be considered "epsem" designs? (circle all that apply)

A) probability proportional to size sampling
B) cluster sampling
C) systematic sampling
D) simple random sampling

Question 6

Of the following samples and populations, which would yield the greatest precision?

A) Sample A = 100; Population A = 10,000
B) Sample B = 50; Population B = 40,000
C) Sample C = 2,000; Population C = 10,000,000
D) Sample D = 1,000; Population D = 3,000,000

Question 7

Which of the following sampling procedures requires a sampling frame of potential respondents before the research can be initiated?

A) multistage sampling
B) stratified random sampling
C) cluster sampling
D) simple random sampling

Question 8

Researchers conducted a survey on young adults' sexual behavior, and found that many respondents did not answer every question. Specifically, it seems like many people chose to not answer the question, "have you ever had sexual contact with a person of the same sex," most likely because of the stigma of same-sex sexual contact and social desirability concerns (but they did not measure people's social desirability bias). Furthermore, 25 percent of respondents refused to answer the question, "have you have had sexual contact with different people on the same day," but it appears that only those people who reported being in a committed relationship (which was measured) refused to answer this question. In the example, what type(s) of missingness are present?

A) MCAR
B) MAR
C) NMAR
D) A and B
E) B and C
F) all of the above

Question 9

June used a multistage sampling to conduct a survey of current political attitudes in a major metropolitan city. First the city was broken into neighborhoods, and then some participants were selected out of each neighborhood to be included in the sample. In this example, when was random sampling employed?

A) when neighborhoods were selected
B) when individual residents were selected
C) when the city was selected
D) A and B
E) all of the above
F) none of the above

Question 10

W. Coyote was very interested in the population of rabbits in the state of Arizona, but for some reason, there did not seem to be a centralized list of names of all the Arizona rabbit residents. However, he did obtain a county map, further divided into neighborhoods, that he thought he might be able to use to estimate the number and type of rabbits in the state. Given the information he has available, what sampling techniques should he use?

A) stratified sampling
B) cluster sampling
C) multistage sampling
D) convenience sampling

Question 11

Cyrus does not like conducting interviews in-person, so he wanted to use phone interviews for data collection. What types of questions or issues should Cyrus focus on to maximize the benefits of phone interviews relative to face-to-face interviews? (circle all that apply)

A) any illicit behaviors people have engaged in (e.g., drunk driving)
B) the prevalence of certain sexually transmitted diseases
C) attitudes of people who are impoverished and without a high school diploma
D) a representative national sample using random digit dialing

Question 12

Adria used stratified sampling to collect data on current attitudes toward American car manufacturers, while Benson used cluster sampling to survey people's attitudes. It turns out that the strata/clusters that both Adria and Benson drew their respondents from were quite heterogeneous. In this case, heterogeneous strata _____(a)_____ the precision of Adria's sample, and heterogeneous clusters _____(b)_____ the precision of Benson's sample.

A) increased; lessened
B) increased; increased
C) lessened; lessened
D) lessened; increased

Question 13

Random sampling is most important for establishing internal validity, while random assignment is most important for achieving external validity.

Question 14

The precision of a population estimate would be improved or increased by which of the following?

A) sampling from a fairly homogenous population
B) smaller sample sizes
C) a larger standard error of the sample mean
D) increasing the sampling frame
E) all of the above

Question 15

The purpose of the finite population correction (fpc) is to:

A) correct for poor precision
B) correct for the fact that sampling is done without replacement
C) correct for an inaccurate standard error of the mean
D) improve the precision of an estimate with small samples
E) none of the above

Question 16

Studies in which every unit in a population has equal probability of being sampled for study, or in which the probability of selection is exactly equal to the sampling fraction, employ_________ sampling methods.

A) cluster
B) nonrandom
C) simple random
D) multistage
E) epsem

Question 17

Using a phone book to create a list of all fast food restaurants in LA, randomly selecting 20 of these restaurants, then interviewing each employee at each of the selected 20 restaurants is an example of:

A) multistage sampling
B) cluster sampling
C) stratified sampling
D) two-phase samplings
E) systematic sampling

Question 18

Say a researcher is interested in studying college students' opinions on admission policies. The researcher wishes to obtain a fairly equal number of first year, sophomore, junior, and senior students, as recent research has shown that older students tend to be less approving of school policies in general. However, there are much fewer older (more senior) students in attendance than younger students. What kind of sampling technique might the researcher want to employ?

A) systematic sampling
B) simple random sampling
C) cluster sampling
D) proportionate stratified sampling
E) disproportionate stratified sampling

Question 19

What is the difference between quota sampling and stratified sampling?

Question 20

Rather than sampling individuals, these sampling methods utilize sampling frames of specific locations:

A) cluster sampling
B) two-stage sampling
C) stratified sampling
D) simple random sampling
E) multistage sampling

Question 21

A psychology professor asking her students to complete a survey for a study she is conducting, and asking them to invite their friends and family members to participate, would be an example of:

A) two-phase samplingcCluster sampling
B) convenience sampling
C) snowball sampling
D) quota sampling

Question 22

__________ are advantageous because they are likely to provide valid data, particularly with sensitive or socially undesirable topics.

A) face-to-face interviews
B) panel surveys
C) time series designs
D) telephone surveys
E) focus groups

Question 23

When data points are not missing at random (NMAR):

A) missing values cannot be predicted or imputed
B) missingness is explained by one or more variables in the dataset
C) more options are available than when data are missing at random (MAR)
D) missingness depends on an unmeasured or unobserved variable
E) None of the above

Question 24

Explain why we do not necessarily need to conduct a census to have a good estimate of a population's scores. Specifically, based on the formula for calculating the standard error of the mean for a sample, how does sample size influence the precision with which a sample approximates the true population value?

Question 25

You are preparing to conduct a survey regarding a new personality characteristic and whether it correlates with other important personality characteristics, but realized you do not really have the time or resources to use randomized sampling techniques, and instead opt for a nonrandom sampling approach. (a) Describe two nonrandom sampling techniques, and (b) provide an example for each type that illustrates how you would eventually obtain your sample of respondents.

Question 26

(a) Discuss two techniques that researchers have used to reduce nonresponse during data collection. (b) Assuming that many randomly selected respondents nonetheless refuse to participate, describe two techniques that researchers can use to reduce the biasing effect that these refusals might have on survey estimates.

Question 27

Describe the following terms, including how they are similar and different: missing completely at random (MCAR); missing at random (MAR); and not missing at random (NMAR).

Question 28

You are preparing to conduct a nationwide survey of people's political attitudes, and are debating about which sampling technique to use. Specifically, you are deciding between proportionate stratified (random) sampling, and multistage sampling. For each sampling technique: (a) what information will you need for your sampling frame; (b) what are the basic steps in selecting who will participate in the survey; (c) how many strata/clusters will you use, and what will those strata/clusters be; and (d) where does randomization occur in the design (or does it occur anywhere)?

Question 29

Describe the relationship between precision and sampling error. How would a smaller sample differ from a larger sample with regard to each?

Question 30

What features distinguish proportionate stratified samples from disproportionate stratified samples? In which research contexts might one be preferred to the other?

Question 31

How can participant nonresponse affect the effectiveness of a sampling strategy? What are some ways researchers can reduce nonresponse?

Question 32

If we want to conduct a representative national survey with approximately 2,000 respondents, with the goal of obtaining a relatively accurate idea of people's opinions regarding a certain national issue, why might it be necessary to over-sample certain groups? For the purposes of our survey, why would we want to necessarily find more Caucasian women, who are in a committed same-sex relationship, with children, compared to any other combinations of demographic characteristics? Why wouldn't we want to find a number of respondents who fit this group that is proportional to the size of the larger population (e.g., if 0.1 percent of people in the population fit this category, why would we want to recruit more than two respondents from this group)?