Sampling and Non-response (W3Y2)

Question 1

How does non-response lead to sampling bias?

Answer 1

If only a small fraction of the randomly sampled people choose to respond to a survey, the sample may no longer be representative of the population.

Question 2

List 2 issues with taking censuses.

Answer 2

• Difficult to complete a census as there always seem to be some individuals who are hard to locate or hard to measure.
• Populations rarely stand still as its constantly changing, so it is impossible to get a perfect measure.

Question 3

What is generalisation?

Answer 3

Distinction between the population of interest and the actual population defined by the sampling frame

Question 4

List the 3 methods of probability sampling.

Answer 4

• Simple
• Stratified
• Cluster

Question 5

List the 3 methods of non-probability sampling.

Answer 5

• Quota
• Convenience
• Purposive

Question 6

How does stratified cluster sampling work?

Answer 6

• First you define the clusters
• .Then you group the clusters into strata of clusters, putting similar clusters together in a stratum
• Then you randomly pick one (or more) cluster from each of the strata of clusters
• Then you sample the subjects within the sampled clusters (either all the subjects, or a simple random sample of them)

Question 7

List 3 advantages of stratified cluster sampling.

Answer 7

• Reduce the error in cluster sampling by creating strata of clusters
• Sample one cluster from each stratum
• The cost-savings of clustering with the error reduction of stratification

Question 8

How is measurement a source of error in sample survey research?

Answer 8

Bad question or instrument design leads to poor measurement validity

Question 9

How is non-response a source of error in sample survey research?

Answer 9

Considering what if not every person answers every question or what if some people decline to take the survey as it leads to inadequate coverage of the population

Question 10

What is item non-response?

Answer 10

An individual skips a certain question on a survey or refuses to answer it.

Question 11

What is survey non-response?

Answer 11

An individual refuses to participate altogether.

Question 12

How do we avoid non response before it occurs?

Answer 12

• Persistence at contact maybe through introductions and endorsements
• Incentives

Question 13

How do we deal with survey non-response?

Answer 13

Re-weighting as some parts of population are under-represented in respondent data so we weight these parts up to compensate for under- representation

Question 14

What is the equation for sampling weight?

Answer 14

Sampling weight = 1/response rate

Question 15

What are the limits of re-weighting?

Answer 15

This only works if we can measure all the factors affecting likelihood of response.