Class_14

Class_14 - Testing Hypotheses about Differences among...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Testing Hypotheses about Testing Hypotheses about Differences among Differences among Several Means Several Means
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The t-test compares independent means TWO AT A TIME . Whenever we encounter a research situation in which we need to compare THREE or MORE independent-sample means, we would have to make multiple t-tests. This is clear from the numerator of the t-test: ( 29 ( 29 1 2 ˆ 1 2 1 2 Y Y Y Y t - - - - = σ μ
Background image of page 2
Statistically, this is undesirable. Recall that each t-test is conducted with a chance of incorrectly rejecting the null hypothesis. The magnitude of this chance is determined when we decide upon an alpha level. With alpha set at 0.05 , there is a 5 percent chance of committing a Type I error each time we perform a t-test. When we perform 100 t-tests , we will INCORRECTLY reject the null hypothesis 5 times . Clearly, the chance of error increases in proportion to the number of t-tests we make.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
That is, each student's score could be predicted by knowing the value of μ (the average exam performance of all PPD 404 students in general) plus knowing how students differ from one another in every respect except their instructor.
Background image of page 6
α would NOT equal 0.0, and our prediction would need
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
In performing an analysis of variance, we do not test the differences among means directly. Instead, we make two different estimates of the
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/04/2008 for the course PPD 404 taught by Professor Velez during the Fall '07 term at USC.

Page1 / 18

Class_14 - Testing Hypotheses about Differences among...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online