Correlation Flashcards
What is correlation?
Looking at how 2 variables are related to each other.
What is the difference between correlation and regression?
- We aren’t making predictors from one to another
- The relationship is symmetrical, unlike regressions.
What statistic do we use to calculate the correlation coefficient?
Pearson’s ‘r’
What does a correlation coefficient show us?
- The strength of the relationship
- The direction of the relationship
What does the p value tell us?
Whether we can accept or reject our null hypothesis.
If p<0.05, we can reject the null hypothesis.
What does a correlation matrix show?
Shows the interrelations between multiple variables - series of correlations.
What is the biggest Correlation Rule to remember?
Correlation is not sufficient evidence for causality between variables.
Correlation coefficients give no indication of the direction of causality.
What could cause a correlation if the 2 variables do not cause each other?
The relation may be due to each varible’s relation to a third.
What are partial correlations?
Correlations that estimate the relationship between two variables X and Y by controlling for Z.
When do we use partial correlations?
When we think there could be a third variable.
What assumptions do you check for when doing a correlation?
- Linearity
- Normality
What are the non-parametric methods of correlation?
- Spearman’s Rho
- Kendall’s Tau
How do we come by correlation?
Using a statistic called covariance
What is covariance?
Measure of how two variables vary together.
It tells us how much each PPs scores differ from the means.
What are residuals?
The difference between the scores and the means.
