Sampling

Question 1

What is sampling?

Answer 1

Process of selecting units from a population of interest to generalize results back to the population.

Question 2

How to generate a sample?

Answer 2

5 steps:

Start with population of interest.

Find sampling frame.

Mirror the image of the population

Sampling techniques to draw sample.

Sampling bias.

Question 3

What is the sampling frame?

Answer 3

A formal and exhaustive list of units or elements that constitute the population.

Must represent every possible unit in the population.

Question 4

How do we ensure external validity based on sampling?

Answer 4

Sample must fairly represent the population. The effect size and direction of sample are similar to that of the population.

Question 5

What is an element?

Answer 5

Basic unit that defines the study population (people or schools).

Question 6

What is the goal of any sampling technique?

Answer 6

To maximize generalizability of the sample to the population.

Question 7

What is sampling bias?

Answer 7

Difference between sample and population that is not attributed to chance.

Systematic difference in sampling such that some members of the intended population have a lower or higher sampling probability than others.

It results in a biased sample of a population in which all individuals, or instances, were not equally likely to have been selected.

Question 8

What is extrapolation?

Answer 8

Process of using sample to make inferences about an entire population. Reducing sampling bias allows for extrapolation.

Question 9

What is random selection?

Answer 9

Any sampling method that allows chance to operate in the selection of units from a sampling frame.

Question 10

What is simple random selection?

Answer 10

Each element has an equal and non-zero chance of being selected.

Question 11

What is systematic random sampling?

Answer 11

Select one random variable as starting point then each sample is selected at an interval.

Does not ensure equal and nonzero chance of selection.

Question 12

What is an interval?

Answer 12

A pre-determined distance between elements in a sampling frame.

Question 13

What is periodicity?

Answer 13

Inherent ordering of particular characteristic within sampling frame. Potential to enhance representativeness of sample.
There is a pattern in the population consistent with the value of k that is chosen in systematic sampling.

Question 14

What is stratified random sampling?

Answer 14

Identification of some variable that may be related indirectly to the research question and this variable is used to divide population into strata.

Then random or systematic sampling.

Proportionate:

Represents overall population and key subgroups.

Disproportionate:

Uses different sampling fractions to over-sample a small group.

Question 15

What does stratified random sampling allow for?

Answer 15

Ensures critical confounding variable is mirrored perfectly between sample and population.

Sample bias for at least one variable is reduced to 0.

Question 16

What is cluster sampling?

Answer 16

Researchers divide a population into smaller groups of clusters. The sampling frame is made up of these clusters. A simple random sample is then obtained.

The cluster is the intended unit of analysis.

Good for policy or organizational BH studies.

Question 17

What is multistage cluster sampling?

Answer 17

Combines cluster with simple random.

Select random clusters (primary sampling unit).

Then employ simple random sampling from each cluster (secondary sampling unit).

Question 18

What is multistage cluster sampling prone to?

Answer 18

Design effect:
When sampling elements are inherently correlated with one another.

Fixed with large number of clusters; they tend to be homogenous which reduces sampling error.

Question 19

What is stratified multistage cluster sampling?

Answer 19

Clusters may be divided into strata before random selection occurs.

Reduces sampling error to 0 for variables that may otherwise introduce sampling bias.

Question 20

List types of probability sampling.

Answer 20

6:

Simple random.
Systematic random.
Stratified random. 
Cluster.
Multistage cluster.
Stratified multistage cluster.

Question 21

What is probability sampling?

Answer 21

Probability sampling is defined as a sampling technique in which the researcher chooses samples from a larger population using a method based on the theory of probability. For a participant to be considered as a probability sample, he/she must be selected using a random selection.

Question 22

What is convenience sampling?

Answer 22

Use pre-existing groups to facilitate recruitment of sample. People in group my be approached and asked to volunteer.

High risk of sampling bias (no procedure for representativeness).

May be differences between preexisting group and population.

Question 23

What is venue-based-time sampling? VDT?

Answer 23

Two-stage sampling that allows researcher to generate a sample with known properties from hard to reach populations.

Improves upon convenience sampling as it interjects a systematic approach to make inferences about the venue-visiting population.

Question 24

What is purposive sampling?

Answer 24

Also known as judgmental, selective, or subjective sampling, is a form of non-probability sampling in which researchers rely on their own judgment when choosing members of the population to participate in their surveys based on the purpose of their research.

Question 25

What is quota sampling?

Answer 25

Match known parameters of population to sample.

Identify characteristics of population to be reflected in sample.

Occurs when the distribution and proportion of certain characteristics in population is determined and the sample selected mirrors it.

Question 26

What is chain referral or snowball sampling?

Answer 26

There is an initial element (seed) that is randomly selected from a population.

Person who qualifies for study and can help chain grow.

PPT is asked to identify others who meet eligibility criteria and asked to PPT (direct facilitation).

OR

PPT asked to refer them to researcher (indirect facilitation).

Question 27

What is homophily?

Answer 27

High degree of similarity between elements in a sample. This bias is of unknown size and direction. Inferences about target population cannot be made.

Question 28

What is respondent-driven sampling?

Answer 28

Combine snowball with probability.

Seed identified and asked to PPT then distribute coupons to target population.

Produces representative sample of hard-to-reach populations.

Respondent-driven sampling (RDS), combines “snowball sampling” (getting individuals to refer those they know, these individuals in turn refer those they know and so on) with a mathematical model that weights the sample to compensate for the fact that the sample was collected in a non-random way.

Question 29

What are steps of VDT?

Answer 29

5 steps:

1) Identify list of all potential venues.
2) Enumeration of days and times and times that will yield most potential PPTs.
3) Create final sampling frame.
4) Random sample drawn from that enumeration will become recruitment calendar.
5) Imaginary swatch pre-determined for asking people to enter study and make sure they have not done so before. Systematic recruitment at venue.

Question 30

What is sampling error?

Answer 30

A sampling error is a statistical error that occurs when an analyst does not select a sample that represents the entire population of data.

The sampling error is the difference between a sample statistic used to estimate a population parameter and the actual but unknown value of the parameter.

Question 31

What is standard error?

Answer 31

Measure of precision.

Measure of accuracy of sample mean as estimate of population parameter.

The standard error (SE) of a statistic is the standard deviation of its sampling distribution or an estimate of that standard deviation.

Question 32

What is standard deviation?

Answer 32

Measure of dispersion.

Measure of the amount of dispersion or variance of a set of values. A low standard deviation indicates that the values tend to be close to the mean of the set, while a high standard deviation indicates that the values are spread out over a wider range

Question 33

List common non-probability sampling techniques.

Answer 33

6:
Convenience sampling.
VDT sampling.
Purposive sampling.
Quota sampling.
Chain or snowball sampling.
Respondent-driven sampling.