Lec 6: Sampling Methods

Sampling Frame

• List of all items in population from which sample will be selected.

• Sampling Frames influence results of an analysis

  • Using different sampling frames can lead to different conclusions

  • You should always be careful to make sure frame completely represents a target population

  • Otherwise, any sample selected will be biased, and results generated by analyses of that sample will be inaccurate

Sampling

• Process by which members of a target population are selected for a sample

• “instant poll” found on a web page, are naturally suspect as such techniques do not depend on a well-defined frame

• The sampling technique that uses a well-defined sampling frame is probability sampling

Clearly define target population

• Well-defined group of people or other entities

• Population sizes can vary

• Considerable thought must be given in choosing target population

• Members must possess characteristics

Acquire accurate and complete sampling frame

• Researchers obtain or construct complete, accurate, and up-to-date list of all units in target population

Choose sampling technique to draw representative units from sampling frame

Drawing sample is relatively simple task but fatal mistakes are made

Obtain sufficiently large sample to represent characteristics of target population

Rule of thumb: >20%

Small pop’n variance → small sample sufficient

Probability Sampling

• Sampling process that considers chance of selection of each item (in target population)

• Increases chance that sample will be representative of target population

• Use probability sampling whenever possible because only this type of sampling enables one to apply inferential statistical methods to data collected

Non-probability Sampling

• Use non-probability sampling when chance of occurrence of each item selected is not known to obtain rough approximations of results at low cost or for small scale, initial, or pilot studies that later will be followed up by more rigorous analysis

Simple Random Sampling

• Probability sampling process where every individual or item from population has same chance of selection as every other individual or item

• Every possible sample size has same chance of being selected as every other sample of that size

• SRS forms basis for other random sampling techniques

  • Random means no repeating patterns – i.e. in given sequence, given pattern is equally likely (or unlikely)