Correlation Exam 2 Flashcards
Pearson correlation coefficient
Measure of the linear relationship between two continuous variables
Pearson correlation coefficient possible values
- 0 < r < 1
- 0 > r > -1
- r = 0
Pearson correlation coefficient: 0 < r < 1
indicates a positive correlation (as one variable increases, the second variable also increases)
Pearson correlation coefficient: 0 > r > -1
indicates a negative correlation (as one variable increases, the second variable decreases)
Pearson correlation coefficient: r = 0
indicates no linear relationship
When using a Pearson correlation coefficient, does the measurements need to be on the same scale?
no
Spearman rank correlatioan
- AKA Spearman Rho Coefficient
- Continuous data can be transformed into ranks, then used to calculate Spearman r(s)
Spearman rank correlatioan possible values
-1 and +1
Phi Coefficient
- Estimates the extent of the relationship between two binary variables
Phi Coefficient possible values
- -1 to +1
* The closer the value is to 1, the stronger the relationship
Kappa Coefficient
- AKA Cohen’s Kappa
- Used to examine inter-rater agreements
- The greater the number, the more agreement there is
Kappa Coefficient possible values
0 to 1
Kappa Coefficient: 0
no more agreement between two raters as can be expected on the basis of chance
Kappa Coefficient: 1
perfect agreement (all observations on the diagonal)
If you have nominal data, which one would you use?
- Phi coefficient (Φ)
- Kappa coefficient (Κ)
If you have ordinal data, which one would you use?
Spearman rho [r(s)], continuous data that are not normally distributed
If you have interval data, which one would you use?
Pearson (r)
If you have ratio data, which one would you use?
Pearson (r)
What kind of data do you use with the Phi coefficient (Φ)?
nominal
What kind of data do you use with the Kappa coefficient (Κ)?
nominal
What kind of data do you use with the Spearman rho [r(s)]?
ordinal
What kind of data do you use with the Pearson (r)?
- interval
- ratio
What is the null hypothesis for the Pearson (r)?
H0: ρ = 0
Kappa Coefficient interpretation
- Lower than 0.4 = poor agreement
- 0.4 to 0.75 = fair to good agreement
- Higher than 0.75 = excellent agreement
