WEEK 9 Flashcards
Pearson’s Correlation (Product moment)
Assumptions:
-data must be interval or ratio level
-data must be normally distributed
- no outliers
-related pairs, linearity between variables
Linearity
A straight line relationship between x and y
scatterplots and linearity
- we assign one variable to x and one to y
- Each person’s score on x and y should be plotted where they meet
Types of Relationships
- Negative Linear relationship
-Curvilinear relationship - positive linear relationship
-zero
The Correlation Coefficient
The strength of the relationship between the two variables: correlation coefficient
+1 is a perfect positive relationship
-1 is a perfect negative relationship
0 is no linear relationship
Common and shared variance
-the correlation coefficient is a ratio between the covariance and a measure of the separate variances
- when two variables are correlated, they share variance
e.g. ice cream sales and the weather 0.8 squared is 64%
- if the shared variance is greater than the unique variance, the correlation coefficient will be high
Partial Correlation
- Calculates what the relationship between two variables would be if we could take away the influence of additional variables
-this is called controlling for a third variable