1. If you have not done so, record your size estimate from quick look. It is a judgment sample.
2. Take another judgment sample, but this time, pick 10 different "viruses" that look representative. Record their sizes and find their average on paper in a table like the one below. Use the show button to show the page without a timer. Reminder: you should be recording the size, in number of oval units for each virus, not their red identification number!
3. The next sample will be a random sample using random numbers. Use a random number generator such as the one on the previous page (click to open in a new window). Get 10 random numbers between 1 and 100. Each number will be the number of the virus you will use in your sample. Find the size of each chosen virus; record their sizes and find the average size. They should all be in the range 1 to 12. A table is always a good way to organize data.
| Sample type count = |
|
|
|
|
|
|
|
|
|
|
|
| Judgment sample n = 1 | |||||||||||
| Judgment sample n = 10 | |||||||||||
| Random sample n = 10 |
Here is the data from a samples I took when I wrote this page.
| Sample type count = |
|
|
|
|
|
|
|
|
|
|
|
| Judgment sample n = 1 |
|
|
|||||||||
|
Judgment sample n = 10
|
|
|
|
6 | 1 | 3 | 2 | 9 | 5 | 9 | 5.9 |
|
Random sample n = 10
|
6 | 6 | 12 | 1 | 1 | 3 | 4 | 1 | 9 | 6 | 4.9 |
#28 Questions for discussion
a) Which method do you think came closest to the actual average
virus size?
b) How would you rank the methods and why?
c) What is the more accurate method, judgment or random?
d) What is the effect of sample size on accuracy?