Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

PMR Modules 2-5 > Module #4 - Sampling > Flashcards

Module #4 - Sampling Flashcards

1
Q

What does a census measure?

A

the entire population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What does a sample measure?

A

Part of a population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Why does market research almost always rely on samples rather than censuses?

A

Censuses are too time-consuming and costly

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

When does using a census make sense?

A

When the population is small

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

True or False? A sample is always an estimate of the population

A

True

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Does sampling error decrease or increase as sample size increases?

A

It decreases

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Why is a sample not necessarily less accurate than a census?

A

Because both samples and censuses are subject to NON-SAMPLING error

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What advantage does a sample have over a census?

A

A sample’s smaller relative size means that it is easier to control and minimize non-sampling errors than in a census.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the 5 steps of the sampling design process?

A

1 - Define the population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Which step of the sampling design process is the most important?

A

Defining the target population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Why is defining the target population the most important step in the sampling design process?

A

Because having the right target population is crucial to answering the business question. “It requires an understanding of the business issue and making sure that the selected sample relates appropriately to the business issue.”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are some typical factors considered when defining the target population?

A

The same marketing factors usually considered, such as:

1) geographic areas
2) demographics
3) psychographics (attitudes, interests, beliefs)

Not sure about these last 2

4) Digital behavior data – recent/real-time data from search engines, recent web site visitors, etc.
5) business-to-business - business and organization firmographics (business type, size, etc)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the danger in defining a target population too narrowly?

A

Limiting the ability of the research to answer key business questions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the danger in defining a target population too broadly?

A

“Wasting” interviews and potential marketing programs on customers who are not likely to use their products/services

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is a “sampling frame”?

A

It is a list of all individuals from which the sample is (“to be”?) drawn.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

How does a sample frame help in creating a sample?

A

It provides a cost effective and efficient method of contacting a representative sample of the target population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Why is the sampling frame often broader than the target population?

A

Rarely is a sampling frame a perfect fit for a target population. A broader frame allows for recruiters to use screeners qualify respondents.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is synchrographic data?

A

“Real-time” data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What causes “sampling frame error”?

A

When the sampling frame is not a perfect representation of the target population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Give 3 examples of sampling frame error.

A

The frame excludes some members of the population.

The frame includes members who are NOT part of the population.

The frame includes some members more than once.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is “coverage” error?

A

It’s when the sampling frame does not include some members of the population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

How can you account for “coverage error”?

A

1) Redefine the target population

2) use weights or sampling methods to make the sample more representative of the target population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is RDD (Random Digit Dial)?

A

Proprietary software that starts with known area codes and exchanges, and then generates the last four or two digits.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What problems does RDD address?

A

People getting unlisted numbers, and having cells rather than land-lines

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

How can RDD samples be expensive?

A

They generate “bad” numbers as well as “good” ones.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is a “targeted” list?

A

A list in which all households share a common characteristic, e.g., buying a certain product.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is the main drawback of targeted lists?

A

The list may not be representative of the target population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Which type of sample would researchers ideally use for 100% of their surveys? Probability or Non-probability?

A

Probability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What is a probability sample?

A

Probability samples are samples in which every unit in the population has a known probability of inclusion in the sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

What is the main advantage of a probability sample?

A

It allows researchers to apply confidence intervals and to project out to a known population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What are the four types of probability samples?

A

1) Simple Random Sample
2) Systematic Random Sample
3) Stratified Random Sample
4) Cluster Samples

32
Q

What is a “simple random sample”?

A

A probability sample in which each element has a KNOWN and EQUAL chance of being included.

33
Q

What are 2 advantages of a Simple Random Sample?

A

1) Most statistical tests assume use of a Simple Random Sample. Other probability samples require special calculations for statistical inference.
2) Another benefit of SRS is that it is easy to understand.

34
Q

What is a “non-probability” sample?

A

A sample that does not involve a random selection procress

35
Q

What are three reasons that “Simple Random Samples” are not a preferred method?

A

1) It’s expensive
2) It can result in very small sample sizes for sub-groups of interest.
3) It may be difficult to find a suitable SRS sampling frame

36
Q

What is a “Systematic” Sample?

A

A sample constructed by picking a random starting point and then taking every nth element after that.

37
Q

How do you determine the “skip interval” for a systematic sample?

A

You find the N by calculating what percent the desired sample is of the entire population.

38
Q

What is a “Stratified Random Sample”?

A

It’s when you divide the population into sub-groups first and then you select a random sample from within the population of each strata.

39
Q

What is Disproportionate Sampling?

A

Disproportionate sampling is sampling strata at a different rate than they occur in the population.

40
Q

What is Proportionate Sampling?

A

Proportionate sampling is sampling strata at the same rate as they occur in the population.

41
Q

When do researchers most often use cluster sampling?

A

For intercept surveys

42
Q

Describe “cluster sampling.”

A

When the researcher divides the population into a large number of mutually exclusive sub-groups (“clusters”) and then only a random sample of the clusters is included in the final sample.

43
Q

What is “area sampling”?

A

A common form of cluster sampling, wherein clusters are geographic areas. The clusters are randomly sampled.

44
Q

What is single-area sampling?

A

When all the households/respondents within the area are in the sample

45
Q

What is multi-stage sampling?

A

When a sample is taken from within a random sample of clusters.

46
Q

What are the 3 types of “non-probability” sampling methods?

A

1) Convenience & Judgement samples
2) Quota sample
3) Snowball sample

47
Q

Who is included in a “convenience sample”?

A

People who are convenient to interview, e.g., people at a mall, or physicians at a medical conference (assuming you’re looking for physicians generally, and not conference-attending physicians.)

48
Q

What is a a “judgement” sample?

A

A convenience sample where the researcher uses their judgement to decide who to interview.

49
Q

What type of research are convenience and judgement samples suitable for?

A

Qualitative and exploratory

50
Q

What is the major limitation of convenience and judgement samples?

A

It is incorrect to generalize the results from these samples to the general population

51
Q

What is “quota sampling”?

A

A special case of judgement sample whereby the researchers sets criteria for the sample with the purpose of making it representative of the target population

52
Q

How are “quota sampling” and “stratified sampling” similar?

A

They both have specified sample sizes for different target groups.

53
Q

How are “quota sampling” and “stratified sampling” different?

A

Stratified samples use random sampling within each strata. Quota sampling uses a convenience sample but adds quotas to achieve appropriate representation of sample characteristics.

54
Q

What is the key advantage of Quota sampling?

A

Less expensive and more feasible

55
Q

What is the main drawback of quota sampling?

A

It’s a non-probability sample. Although reasonable quotas may be set for some aspects of the population, the sample may not be representative of the population on other variables.

56
Q

What is snowball sampling?

A

When the researcher builds a sample by asking for referrals from other respondents

57
Q

When is snowball sampling useful?

A

When the target population is hard to locate

58
Q

Why do researchers always face a trade-off between sample size and costs?

A

Because sample is one of the major sources of cost for a study.

59
Q

Why are larger sample sizes more desirable?

A

Because a greater number of responses reduces the range of sampling error.

60
Q

What are the 3 key issues to consider when determining sample size?

A

1) The importance of the decision to be made
2) The difficulty and expense of gathering data from certain respondent groups.
3) the need to examine the results by subgroups

61
Q

What is a confidence interval?

A

A range of values, centered on the sample estimate, that is known to contain the true population value with a given degree of confidence

62
Q

When is the error range of a sample affected by the POPULATION size?

A

When the sample represents a substantial proportion of the population, e.g., 5% or more

63
Q

What does “precision” mean with respect to sample size?

A

A measure of how accurate you need the estimates to be

64
Q

What does “power” mean with respect to sample size?

A

How much do you need to identify differences when they exist?

65
Q

What is incidence?

A

Proportion of the population who meet the criteria to qualify for the survey

66
Q

What is “starting sample size”?

A

It is the number of possible respondents needed to eventually reach the desired number of completions

67
Q

What factors affect starting sample size?

A

1) Incidence rate
2) contact info accuracy
3) refusal rates

68
Q

When is “non-response” particularly problematic?

A

When different groups in the population respond at different rates

69
Q

What are 2 ways to adjust for non-response?

A

1) adjusted initial sampling rates

2) Weighting

70
Q

Why are high non-response rates a problem?

A

1) It makes research more costly and time-consuming

2) Low response rates call into question whether the responders are actually representative of the population

71
Q

How can researchers improve response rates?

A

1) telling respondents the importance of the survey
2) Incentives
3) Well designed & short surveys

72
Q

What are 4 key concerns when using Internet sampling?

A

1) Non-representative sample
2) Low response rates when not using panels
3) Respondent fraud
4) Over-representing frequent Internet users

73
Q

How is “self selection” affected by very low response rates?

A

It is an even greater problem than otherwise

74
Q

How might frequent internet users be over-represented in an online survey?

A

Frequent internet users tend to respond to surveys quicker/earlier.

75
Q

How do you avoid having your sample have too many early- responding frequent Internet users?

A

By keeping the field period open long enough for later responders to have a chance to get involved.

76
Q

Why would a researcher prefer to use a probability sample for all studies?

A

To assure the most representative and generalizable sample possible

PMR Modules 2-5 (4 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions