4: Chapter Sample Surveys in the Real World
Sampling and statistics seem simple, but some problems can arise. Example: Prediction poll mistake of the 1948 presidential election that proclaimed Thomas Dewey as the winner over Harry Truman.
Sampling Errors
Sampling Errors are errors caused by the act of selecting a sample. They cause sample results to be different from the results of a census. (50)
Improper sampling techniques used. Undercoverage occurs when some groups in the population are left out of the process of choosing the sample. (51) Random Sampling Error: This results from chance selection in the simple random sample. (50) The error is due to chance. The best we can do to control this is to select a large sample. Margin of error includes only random sampling error.
1
Nonsampling Errors
Nonsampling Errors are errors not related to the act of selecting a sample from the population. They can be present even in a census. (50)
Nonresponse (Missing Data): Refusal to answer survey, subject is not available for survey Response Errors: Subject may lie or remember incorrectly, subject may not understand question Processing Errors: Math errors, coding data incorrectly Effects of Data Collection Procedure:
Wording of the question, Timing How the survey is administered: Mail (low response), telephone, personal interview
How to Live with Sampling Errors
Substitute other households for nonresponders
Can help reduce bias
Weight the responses
Example: Number of males and females in the sample may differ, weighting can account for differences.
Can also account for differences by race, age, household size
Helps correct bias Increases variability
2
More Complex Sample Designs
Sometimes a strict simple random sample is difficult to obtain. Therefore, we need to find other ways of randomly selecting units for a sample. Multistage Sampling Design: This type of sampling is used to select a sample from a very large population where certain groups and subgroups are available.
Example a of Multistage Sample
Suppose we wish to obtain a sample of people in the United States. 1) Randomly select a few states (SRS) 2) Randomly select a few counties (SRS) 3) Randomly select a few neighborhoods (SRS) 4) Randomly select people (SRS)
3
More Sampling Designs
Cluster sampling: Divide population into clusters. Select one or more cluster and include everyone in sample.
Example: SC has 46 counties. Select 5 counties at random, use all household in each selected county as sample.
Systematic sampling: Take every nth item from the sampling frame.
Example: Take every 10th towel from a roll of paper towels
Stratified Random Sampling Design (p. 58)
Step 1: Divide the sampling frame into groups of units called strata.
The strata are chosen using some characteristic of the units already known (i.e.: race, gender, location, etc.). The strata are chosen to be of special interest in these groups within the population. (Note: strata is plural, stratum is singular.)
Step 2: Take a separate simple random sample in each stratum and combine these to make up the stratified random sample.
4
Example: Males and Females
The class has been separated into males and females. Ther...