The Logic of Sampling. Key Sampling Concepts Sampling (two types) Element Population Sample Sampling Frame Representative Sample.

The Logic of Sampling

Key Sampling Concepts

• Sampling (two types)• Element• Population• Sample• Sampling Frame• Representative Sample

Application of Sampling Terms

Types of Samples

• Probability -- Strictly following two rules.• Non Probability -- Failing to follow the two

rules

Types of Probability Samples

• Simple Random Sample (SRS)• Systematic Random Sample• Stratified Random Sample• Cluster Sample

Simple Random Sample

• Every element has an equal chance of selection

• No element can be selected more than once

Example of a good random sample outcome

Simple Random Sample

• Every element has an equal chance of selection

• No element can be selected more than once

Systematic Random SampleSampling Frame

Nth ElementA simple random sample employing a sample

frame.

Stratified Random Sample

Cluster Sample

Non probability samples

• Convenience (available to researcher)• Snowball (available connections)• Quota (stratified without randomness) • Informant (case study/social history)• Focus Groups

What’s the difference?How important is the difference?

Probability samples can be generalized to a population; while non-probability samples cannot.

Non-probability offer an in depth understanding and are most often: “I don’t know what I am seeking until after I find it.”

Following is an illustration:

Overview of Sample ProblemsHite, S. (1987). Women in Love: A Cultural Revolution in Progress. NY: Alfred A. Knopf.

Laumann, E. O., Gagnon, J. H., Michael, R. T. & Michaels, S. (1994). The Social Organization of Sexuality: Sexual Practices in the United States. Chicago: University of Chicago Press.

Sample Size Selection

The problem with the following formula:It is calibrated for dichotomous data. The sample size will increase with the number of options given to the subject.

Do NOT Forget!!!!

Regardless of what formula one uses,

always increase the sample size by 20%.

Excel

• RANDBETWEEN – Returns a random number between the numbers you specify.

• RAND – Returns a random number greater than or equal to 0 and less than 1, evenly distributed (changes on recalculation).