Class Notes (1,034,982)
CA (593,235)
U of G (29,258)
STAT (236)
Lecture 17

# STAT 2230 Lecture Notes - Lecture 17: 5,6,7,8, Statistical Unit, SunscreenPremium

3 pages67 viewsWinter 2017

Department
Statistics
Course Code
STAT 2230
Professor
Lecture
17

This preview shows half of the first page. to view the full 3 pages of the document.
Day 17: Paired design [Feb 15]
- assumptions of independent samples t-test:
- observations are independent
- observations are normally distributed
- dispersion is equal to the two groups (‘equality of variances’)
- ////CHART + a few other pages
- paired design: comparing two groups when groups are paired
- in a paired design, each observational or experimental unit in group 1 is
paired or matched with an observational or experimental unit in group 2
- the members of a pair have more in common with each other than with
members of another pair
- paired designs make it easier to detect differences between groups when
they exist
- only concerned with differences between groups within a pair
- differences among pairs are ignored
- example:
- researchers are interested in comparing fish species directly in
lakes before and after heavy metal contamination
- comparing patient weight before and after hospitalization
- testing effects of sunscreen applied to one arm of each subject,
compared with a placebo applied to the other arm
- testing effects of smoking in a sample of patients, each of which is
compared with a non-smoker closely matched by age, weight, and
ethnic background
- paired t-test
- d = difference between each pair of measurements
- degrees of freedom (df) = number of pairs - 1
- ////CHART
- midterm review notes:
- multiple choice and short answer
- 29 questions
- know when you would use pt vs. 1-pt vs. 2xpt, etc.
- answers to the sample midterm will be posted after class!
- example:
- 1, 2, 3, 4, || 5, 6, 7, 8
- median is 4.5
- 1, 2, | 3, 4, | 5, 6, | 7, 8
- interquartile ranges
- Q1 = 1, 2
- Q2 = 3, 4
- Q3 = 5, 6
- Q4 = 7, 8