STA 3024 Introduction to Statistical 2

1
STA 3024Introduction to Statistical 2

• Review of Chapter 10

2
Key Formulae
3
(No Transcript)
4

• From
• P.534

5
Examples from the Book

• Aspirin and heart attack example, p.469.
• Effect of cell phone on driving safety example,
  p. 488.

6
Aspirin Example p.469 (Harvard Medical School
five year randomized trial)
Can we claim that aspirin reduces heart
attack? From the nature of this trial, we feel a
two-sided test is more appropriate. Thus we may
use confidence interval to draw the conclusion.
7
(No Transcript)
8
(No Transcript)
9
Cell Phone Use and Driving Example p. 488 (U. of
Utah case-control study)

• Reaction time to red light in milliseconds.

Can we conclude that cell phone delays the
reaction time (on the average)? The test can be
one-sided or two sided depends on the initial
intention. We follow the books two-sided test.
10
(No Transcript)
11
(No Transcript)
12
Assuming Equal Variance (p. 495)
13

• The box plot show that the cell phone group has
  an outlier, with a very slow reaction.
  (approximate 4? higher than the mean.) This datum
  can bias the mean and the standard deviation
  estimates badly.
• When this data is eliminated, the new analysis is
  given in the middle of page 490. The results is
  very similar. (p-value0.015)
• In practice, eliminating data is a serious matter
  in scientific investigation. It has to be done
  with valid reason other than the value itself.

14
Using Software SPSS for data analysis(Instruction
s given in the class website)

• Classroom demo
• Two sample t-test Exercise 10.20 eBay auctions
  (p.492)
• Paired t-test Anorexia example (p.433)