More Sampling Methods
Chapter 4 - Day 2 - Lesson 4.1
Describe how to select a sample using stratified random sampling and cluster sampling, distinguish stratified random sampling from cluster sampling, and give an advantage of each method.
Activity: How Much Do Fans Love Justin Timberlake?
For many years, I used the Cry Me a River Problem to teach this lesson. But I think this context is better because anything with Justin Timberlake is better.
As an extension to this Activity, we asked students how we could take a cluster sample. They decided that we should use the columns as clusters. Randomly select one column and then sample all four sections in that column.
Use this song as students walk into the room.
Make the three dotplots on a poster board and then save the poster board to reference later in the year. Each dotplot is a sampling distribution (which we don’t formally study until Chapter 7). Students also saw sampling distributions in the Beyonce Activity.
Be sure to say Bye Bye Bye at the end of the class period.
There are actually three specific ideas we want students to get out of this activity:
A stratified random sample can be more effective than a simple random sample because it reduces the variability in the sampling distribution.
When choosing the variable to use for stratifying (row or column), pick the one that most reduces the variability in the sampling distribution or the one that is most strongly associated with the response variable (enjoyment).
Stratified random samples take an SRS from each strata, while a cluster sample takes a random sample of clusters then uses all individuals in the selected clusters.