Test of Relationships - Correlation (r) Essential Reading - Andy Field, Chapter 8 (Correlation); Chapter 5 (Exploring data using graphs) What is correlation? An indication as to the extent to which two variables covary. It tests the extent to which two variables are related A positive correlation is where variable X increases by one unit, so does variable Y A negative correlation is where variable X increases by one unit but variable Y decreases When to use correlations You use a correlation test any time you want to examine a relationship between variables of interest Correlation coefficient (r) -1.00 0 +1.00 Strong negative relationship Strong positive relationship No relationship There could be genuine relationships but relationships do not mean causation. There may be a third (fourth etc.) variable that explains the link, called confounding variables and extraneous variables. Bi-directional links is where the DV may cause the IV or vice versa. Scatterplots The graphs are used to draw relationships after using correlation X - Predictor variable (IV) Y - Outcome variables (DV) Each point represents a participant. It is positioned on the graph depending on the participant's score on each of the two variables Line of best fit Y = a + bX A is the constant and B is the coefficient The degree of a relationship R = . 90 is a strong relationship; R = . 70 is a moderate relationship; R = . 00 is no relationship
You may want to check for outliers