page4

..

original

When all have recorded their data, take a look at the patterns. Think of each pack as a class of diverse students. What are the differences among each pack. Would any one of the packs be representative of all the packs? What if, by chance alone, your sample consisted of the pack most different from the others? Would you think that the results gleaned from another pack would represent your pack? How many of these packs should we choose if we want to represent the total amount of students? The number of people in our sample also has major implications for the validity of our study. How does the sample size influence the study?
 * ASSIGNMENT 2: (Posting deadline November 14. **


 * POSTING 2 COMMENT ON PATTERNS YOU SEE AND HOW THIS RELATES TO THE INFORMATION PRESSENTED ABOUT SAMPLING IN THIS SESSION. CHAT WITH OTHES ON YOUR POSTING PAGE ABOUT THEIR CONCLUSIONS AS WELL. **

Inferential statistics is based on the idea that the results of your study can be generalized to other groups. If the other groups are not made up of the same kinds of students your class had, then the results may or may not represent your class. We will explore this idea later as we want to make sure that our results are not due to a sampling phenomenon. That is what are the chances that our results represent a fluky sample or bag of m&ms. Too few students, as in the case with the M&Ms, will not give us a true picture of a phenomenon. The power point on sampling reveals ideal numbers of subjects for different kinds of studies. For action research we don’t always have the luxury of finding a perfect sample size. We might just use a convenience sample described later

.