Session 5 - Correlation Flashcards
Correlation Coefficient
Used to measure the strength of relationship between 2 continuous variables
Displayed using scatter diagrams - each point = one subject
Calculation of Correlation Coefficient
Product of Differences from the Mean of 2 variables
For each observation - find difference from the mean, multiply the differences for the two variables, add them together
= ‘sum of products about the mean’
Pearson’s Correlation coefficient
=r
= product moment correlation coefficient
Min -1, Max +1
No linear relationship r=0
Correlation coefficient guidelines - rule of thumb
r = 0.0 No correlation
0.0 < r < 0.2 Very weak correlation
0.2 ≤ r < 0.4 Weak correlation
0.4 ≤ r < 0.6 Moderately strong correlation
0.6 ≤ r < 0.8 Strong correlation
0.8 ≤ r < 1.0 Very strong correlation
r = 1.0 Perfect correlation
Assumptions for Pearson’s correlation coefficient
One or both variables are normally distributed (large deviations = unreliable p)
Observations are independent
Linear relationship
Two continuous variables
What if assumption or normality not met for Pearson’s
Use alternative correlation coefficient
ie Spearman’s Correlation Coefficient
Works by first ranking observations
Then calculate product moment correlation of the ranks
Denoted as ρ or rs