Pearson’s correlation coefficient – how to interpret it?

Things to remember

Pearson’s correlation coefficient r measures the strength and direction of a mutual relationship between two continuous variables.

negative values of r = negative correlation (e.g. r = -.342)
positive values of r = positive correlation (e.g. r = .512)

The r closer to 1 or -1, the stronger correlation
Coefficients r close to 0 represent a weak correlation
If the p-value is below or equals 0.05 (sometimes 0.01)  the correlation is statistically significant

Changing the p-value from 0.05 to 0.01 reduces a Type I error

A statistically significant correlation suggests a reliable relationship, not a strong or weak relationship
The bigger the sample the bigger chance the correlation becomes significant.

Strength of correlation

0.1 – 0.3 weak/small correlation
0.3-0.5 moderate/medium correlation
0.5 -1.0 strong correlation

Examples

In the example above, there was a positive moderate correlation between reading and writing (r (62) = .425,  p < 0.01).
Children who are better readers tend to be better writers.
This correlation was statistically significant since the p-value was less than 0.01 (p < .01).
The number following r in parentheses signified the degrees of freedom (df), which are related to the sample size.
For a correlation, the degrees of freedom is N 2. The sample size (N) was 60.

 

Follow by Email
Facebook
Facebook