Stat 100

1
Stat 100

• Jan. 9

2
To Get Started

• Read Chapter 1, Try Exercises 1-3,5,6, 9
• Optional for now skim Chapter 5

3
Question 1
Suppose you are designing a study to compare two
different methods for improving arm strength.
One group of students will use one method and
another group use the second method. How would
you assign students to groups in the first place?

4
Key Terms

• Response Variable main outcome variable of
  interest
• Explanatory Variable variable that explains or
  predicts the response variable
• Randomized Experiment comparative study in
  which researcher assigns different tasks to
  participants

5
For Question 1

• Response improvement in arm strength
• Explanatory method used
• Its an experiment methods are assigned
  randomly to the participants

6
Question 2
For each of 10,000 women, a medical researcher
records blood pressure and whether or not the
woman uses oral contraceptives. The average blood
pressure is about the same for women who use oral
contraceptives and women who dont use oral
contraceptives? Is this evidence that oral
contraceptive use does not increase blood
pressure? Why or why not?
7
Observational Study

• Observational Study essentially a survey,
  investigator does not assign any tasks to
  participants
• Researchers did not assign women to take oral
  contraceptives or not only surveyed women about
  this

8
Variables of interest -

• Response blood pressure
• Explanatory use of oral contraceptives or not

9
Predictors of Blood Pressure?

• Diet
• Genetics
• Job stress/ stress in general
• Weight
• Exercise
• Age

10
Another Key Term

• Confounding variable a variable related to both
  the response and explanatory variables

11
For question 2-

• Age was a confounding variable
• Blood pressure increases with age
• Oral contraceptive users were generally younger
• So the younger group (users) had about the same
  blood pressure as the older group (non-users)

12
Question 3
Suppose that we want to learn about the
relationship between sun exposure and the risk of
skin cancer. How might we gather data to study
this problem?
13
Another Key Term

• Case control study survey or observational study
  in which investigator samples cases and also
  samples controls
• Cases people with certain condition (like skin
  cancer)
• Controls people without the condition