Sampling- Lab Assignment

 

The Logic of Probability Sampling

1.   To learn about the logic behind probability sampling, go to www.learner.org/exhibits/statistics/ . Answer the poll and then follow the directions for an interactive review of sampling.   Be prepared to summarize the site and what you learned by using the site.  Why do they say that the poll does not represent a statistically valid random sample?  See 20 questions a journalist should ask about poll results to find out the answer http://www.ncpp.org/node/4/#10

(Please wait for everyone to finish before moving on so we can discuss this as a group.  If you finish early, scroll through the other questions/answers on this site as this critical assessment of research is similar to what you’ll be doing on the exam).

 

Sampling Designs

  1. Suppose you wanted to study leadership styles in clubs and organizations at Winthrop.  If you wanted to generalize to all clubs and orgs here, you’d need a probability sample.  That requires getting a list of all clubs and orgs.  How could you find that?  Fortunately for you I’ve already done that part for you (come get a handout from me).  After reading about different sampling designs, conduct a systematic random of clubs and organizations from those listed in the handout.    

 

    1. List cases in the population in a random fashion and number each case (we will make the assumption that an alphabetical listing is sufficient and does not introduce bias)
    2. Determine your desired sample size.
    3. Select the first case at random.
    4. Select every kth case thereafter to reach the desired sample size. (Circle all selected cases).

 

Tip: If you have a population of 10,000 and want a sample size of 200, select every 50th case.

How would you conduct a stratified sample of these clubs and organizations?  Why might you do this?  Which sampling method do you think would be better for the purposes of your research?   Suppose you wanted to know more about what type of people join clubs and organizations but there is no master list of all individuals in clubs and orgs, what kind of sampling design could you employ?

-Think about your answers to each of these questions and we will discuss them together.