Sample DefinitionsSample Types
Rules for Simple Random Sampleand
Sample Size Selection
Basic Terms• Sampling• Element• Population• Sample• Sampling Frame• Representative Sample v. Bias SampleRules for getting a good representative sample• Every element has an equal chance of being
selected.• No element can be selected more than once.
Types of Sampling
•Probability
•Non-probability
Probability Sampling• Simple Random Sample (SRS)
• Systematic Random Sample
• Stratified Random Sample– (Proportional Stratified Random Sample)
• Cluster (Random) Sample
Non-Probability Sampling
• Availability Sample• Purposeful/Judgmental • Snowball• Quota • Informant• Focus Group• Criteria
Rules for a Collecting a Simple Random Sample (SRS)
• Every subject within a population must have an equal chance of being selected.
• No subject within a population may be selected more than once.
Note: Violation of either one of these rules can produce false generalizations. In addition, compliance with these rules is very expensive. For example, the standard fee to conduct a national phone survey is $1,500 per minute. Because of this expense, social service agencies rarely collect adequate samples. Therefore their results cannot be trusted.
Sample Size Selection
• The problem with the following formula:• It is celebrated for dichotomous data. The
sample size will increase with the number of options given to the subject.
One More Rule
There are many formulas for determining a sample size. Regardless of what formula you use, the rule of thumb is, “increase your sample size by 20%.”
Check out the following
• http://stattrek.com/Lesson3/SamplingTheory.aspx?Tutorial=Stat’
• http://www.youtube.com/watch?v=Qo2K46Hr3lc
• http://www.youtube.com/watch?v=g9H_XGqmvEs