RxC contingency tables
Home > Preview
The flashcards below were created by user
on FreezingBlue Flashcards.
What makes an RxC contingency table?
When R denotes rows and C denotes columns (duh)
For RxC tables, how is the null expressed?
- P(Help|0) = P(Help|1) = P(Help|4)
- This is a statement about probabilities in a population
How do you calculate the probabilities for a cell?
Divide the value by the total for that column
How are the expected values calculated?
Row total x column total over grand total
When we collapse over categories?
When it makes theoretical sense to do so
What do we do when there is no clear IV or DV?
We provide a measure of association rather than conditional probabilities
What is the formula for the phi coefficient?
What are the phi ratings and who suggested them?
- 0.1: Small effect
- 0.3: Medium effect
- 0.5: Large effect
What is Cramér's coefficient?
- A more general form of the phi coefficient used for RxC tables
- This was coined by Harald Cramér
What is the formula for Cramér's coefficient?
- N: Total sample size
- k: The smaller figure between R + C
What are the effect sizes for 2 (df = 1) tables?
- Small: .10
- Medium: .30
- Large: .50
What are the effect sizes for 3 (df = 1) tables?
- Small: .07
- Medium: .21
- Large: .35
What are the effect sizes for 4 (df = 1) tables?
- Small: .06
- Medium: .17
- Large: .29
What are the hypotheses for correlation?
What are the hypotheses for probability?
Home > Flashcards > Print Preview