Sample Surveys

1
Chapter 12

• Sample Surveys

2
Samples

• Population
• Vs. Sample
• What is Bias?
• Samples that dont represent every individual in
  the population fairly.
• Any systematic failure of a sample to represent
  its population

3
How do we select the sample?

• Randomly!
• Makes sure that on average, the sample looks like
  the population.
• Does size matter?
• Not really, so long as it is not too small.
• Need enough soup to taste.

4
Should you do a Census?

• If we sample everyone in the population, weve
  done a census.
• Why are they not done more often?
• Twinkies
• Cost
• Time
• Even if we do one, still may have problems
• 2000 Census story

5
How to draw the sample?

• Simple Random Sample
• Like drawing names from a hat
• Stratified Sampling
• Helps insure against a possible
  non-representative sample
• Cluster Sampling
• Systematic Samples

6
Dangers and Disasters

• There are lots of things that can go wrong with
  samples.
• Sources of bias.
• Voluntary Response Sample? Voluntary response
  bias
• Convenience Sample
• Bad Sampling Frame

7
Undercoverage vs Nonresponse

• Undercoverage in selecting the people to be
  sampled, some proportion of the population is not
  sampled at all, or smaller representation in
  sample than population
• Example Call people at home between 10am-2pm.
• Nonresponse People chosen for a survey may not
  respond.
• Those who choose not to respond may be different
  than those who do.

8
Other sources of bias

• Really long surveys
• Non-objective wording
• GP vs environmentalists in Brunswick