Agenda

1
Agenda

• Buy materials
• Go over syllabus
• Preview course
• Take photographs

2
Lecture 0

• Introduction

3
Statistical questions

1. What is typical?
2. How much variety is there?
3. Who exactly are we talking about?
4. How certain are we?
5. What should we compare this to?

4
Proof by example

• Proof by example
• Often used in journalism and politics
• People are trying to sue McDonalds for making
  consumers fat.
• chain restaurants must be protected from
  frivolous lawsuits
• Physician in Las Vegas closed his obstetric
  practice because of high malpractice premiums
• Malpractice verdicts and insurance rates must be
  capped
• I missed my exit
• The road sign should be moved.

5
1. What is typical?

• Response to proof by example
• Get beyond the single case to evaluate whether
  theres a broader problem.
• What proportion? Whats the average?
• Restaurant examples
• What proportion of consumer tort cases involve
  obesity?
• What proportion of those cases go to trial?
• Whats the average verdict?
• Medical examples
• Whats the average malpractice insurance premium?
• What proportion of medical practices close each
  year?
• What proportion close because of insurance?
• What proportion of expectant mothers cant find
  an obstetrician?

6
Proof by average

• The average temperature in San Antonio is 68.6o
• The average malpractice premium is about 10K
  after taxes
• Only .05 of state lawsuits end in punitive
  damages

7
2. How much variety is there?

• Response to proof by average
• Get beyond the average
• How extreme can it get?
• How often does it get that extreme?
• The average temperature in San Antonio is 68.6o
• But summer days often go over 100o
• The average malpractice premium is about 10K
  after taxes
• but its twice that in obstetrics and surgical
  specialties
• and higher in certain counties
• or if you have a history of malpractice
• Only .05 of state lawsuits end in punitive
  damages
• but when damages are awarded they average over 1
  million

8
Everybodys doing it

• Presidential election, 1936
• Literary Digest poll
• Got 10 million names from lists of car and phone
  owners
• Mailed 10 million questionnaires
• Got 2.3 million responses
• Results 57 favor Landon (R), 43 favor
  Roosevelt (D)
• Election result

9
3. Who exactly are we talking about?

• Response to everybodys doing it.
• Who exactly was the Literary Digest talking
  about?
• Car and phone owners
• The wealthy, in 1936
• Who should they have been talking about
• US voters
• a lot of them poor, especially in the Great
  Depression

10
4. What should we compare this to?

• Answer to Is that a lot?
• Average malpractice premium is 10K
• What should we compare this to?
• Average physician salary 205,700
• after taxes and expenses
• including the expense of malpractice insurance
• McDonalds coffee case was settled for 600,000
• What should we compare this to?
• McDonalds coffee sales? gt 1 million / day

11
5. How certain are we?

• 46 of US voters are leaning toward Kerry
• Margin of error /- 3
• Were certain about the 1000 voters we talked to,
  but not about the others
• Theres no significant difference between
  voting for Kerry and Bush
• Theres a difference, but were not sure which
  direction

12
Overview of course
Lectures Question Topics (cumulative)
1-3 Organizing and displaying data
4 Whats typical? mean (aka average), trimmed mean, median, mode
5 How much variety is there? variance, standard deviation, etc.
6-7 Who are we talking about? sampling
8-9 How certain are we? confidence intervals
Midterm
10-11 How certain are we? hypothesis tests
12-15 What should we compare this to? tests and intervals for 2 or more groups
Final