Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
14 Cards in this Set
- Front
- Back
|
Continuous Ordinal Variables
|
Ordinal data that has many possible scores.
|
|
|
Collapsed Ordinal Variable
|
No more than 5 or 6 values or scores.
|
|
|
Gamma (G)
|
PRE Measure of association used for collapsed ordinal-level data.
Values range from (-1 to 1) which indicated strength and direction. Value can be multiplied by 100 to get the percentage fewer errors that would be made if we predicted the order of pairs of cases on one variable from the order of pairs of cases on the other. |
|
|
Spearman's rho (rs)
|
Measure of association for continuous ordinal-level data.
|
|
|
Computing Ns
|
For Ns Multiply the cell by the sum of all cells below and to the right of that cell. Repeat this for all cells and add totals to get Ns.
|
|
|
Computing Nd
|
For Nd Multiply the cell by the sum of all cells below and to the left of that cell. Repeat this for all cells and add totals to get Nd.
|
|
|
Gamma as Symmetrical
|
It does not matter which variable is identified as independent.
|
|
|
Strength of Relationship for Ordinal-level Data
|
weak: 0.00-0.30
moderate: 0.30-0.60 strong: greater than 0.60 |
|
|
Computing Spearman's rho (rs)
|
1)Sort data from high to low (by independent variable)
2)Rank the x and y variables (average their ranks if they are tied) 3)Subtract rank on y from rank on x |
|
|
Rho Squared (rs²)
|
A PRE measure of association that when multiplied by 100 tells us the reduction in errors when predicting the rank of X on Y when the rank of Y is taken into account.
|
|
|
Assumptions for Gamma and Rho
|
1)Random Sampling
2)Ordinal-level of measurement 3)Sampling distribution is normal |
|
|
Null Hypothesis for Gamma and Rho
|
H0: ϒ=0.0 (H1: ϒ≠0.0)
H0: ρ=0.0 (H1: ρ≠0.0) |
|
|
Sampling distribution for Gamma
|
Z distribution for 10 or more samples.
|
|
|
Sampling distribution fro Rho
|
t distribution for 10 or more samples.
|
