Here is an example of how you can use the TI-83 to construct a random sample and a stratified sample

Suppose you want to take a sample of 5 students from a class of size 17 students. Here are some ways to do it

Before you begin using the random number generator on the calculator do the following

1. Enter a random integer (I use the current month day combination - for example 828 for August 28)

2. Hit the STO button

3. Hit Math, left arrow and 1 for rand

Here are some calculator screens to help you:

You should use this sequence of steps whenever you turn on the calculator and want to use randInt(

Now back to the problem. Give each student a number (usually from 1-17 works just fine). Type the following keys on your TI-83:

MATH, left-arrow key, 5The screen you should see is illustrated below:

Now type in 1 , 17 , 5 and a ). Note this is starting value followed by ending value followed by number in sample. The screen you should see is illustrated below

Now hit enter and you will see a screen similar to the following (note your values may be different)

This tells you which students to choose in your sample (you would choose students numbered 5, 10, 11, 17 and 6 ). If you get a value repeated, just hit enter again and take the first non-repeated value as your next value and so on.

Suppose now that you wish to select a sample of 10 students from a class of 30 students and you wish to keep the proportion of women and men in the sample the same as are in the class. Suppose the class has 6 men and 24 women (20% men and 80% women). Since 20% of ten is 2 and 80% of ten is 8 you want 2 men and 8 women in your sample. One way to do this is to number the women in the class from 1 to 24 and the men from 1 to 6, now using a process similar to the one used above, we can get a random sample of 8 women and 2 men. Here are the appropriate calculator screens:

So in this case we would choose men numbered 1 and 4 and the women numbered 19,18,23,2,16,24. You should hit enter again after the women's values to get 2 more.

1. A hospital is interested in running an experiment on diabetes patients to determine if a new drug is effective. They have 40 patients volunteering for the program. They wish to give 20 of these patients the new drug and 20 a placebo. Discuss how you would decide which 20 patients receive the drug and which receive the placebo using the TI-83 as illustrated above
2. You want a sample of 100 students from a college of 5000 students and you want the sample to have the same racial proportions as the student body. The racial percentages of the current student body are 60% white, 20% Hispanic, 10% African-American, and 10% Asian. Describe how you could use the TI-83 calculator and the methods discussed above to construct this stratified sample

1) Number the patients from 1 to 40. Using your TI-83, Hit Math, left arrow and 5 to get randInt(. Now enter 1 a comma 40 a comma 20 and a ). Hit enter to get your 20 random numbers. If any values are repeated again hit enter again to get another 20 and select the numbers you need until you have 20 DIFFERENT random numbers

2) Since 60% of 5000 is 3000, there are 300 white students at the school. Since 20% of 5000 is 1000, there are 1000 Hispanic students at the school. Since 10% of 5000 is 500, there are 500 African-American and 500 Asian Students at the school. Number the white students from 1 to 3000, number the Hispanic students from 1 to 1000, Number the Asian students from 1 to 500 and number the African-American students from 1 to 500. To get the samples you need to select 60 white students, 20 Hispanic students, 10 Asian students and 10 African-American Students. So Hit Math left arrow and 5 to get randInt(. Now enter 1 a comma 3000 a comma 60 and a ). Hit enter to get your 60 random numbers these are the students you should select. If you need to hit enter again because some random numbers are repeated you should do so. Do a similar process with the other racial groups to get their random sample