Week4 – Examples

T-tests seem to be the test that most people like to migrate towards.  As nice and easy as a t-test may be, remember that you can ONLY look at 2 groups, if you have more than 2 groups, you need to conduct an ANOVA or F-test.  You CANNOT nor should you ever run several t-tests, by doing this you are increasing your comparison-wise error!  Essentially increasing the chance of finding something significant when it doesn’t exist because you are asking the same question over and over to the same data.

With this in mind, you are presented with the following data and are asked to conduct a t-test. Which t-test will you conduct and how will you enter the data into SAS and what coding will you use?

You’ve been approached by a local grocery store and are asked to see whether the average age of the shoppers is different during the day than at night.  Here is the data you are presented with:

One more t-test example

You’ve been told that you need to ensure that the average height measurements of your purebreed Soft-coated Wheaten Terriers meet the CKC standard of height of 47 cm for the males and 44cm for the females.  You have 5 males and 12 females at your kennel with the following measurements:

Males: 38, 39, 49, 42, 46
Females: 40, 41, 44, 46, 45, 43, 42, 42, 45, 46, 47, 44,

What t-test do you conduct and what is your SAS coding?