4.2.7 About Pearson’s Correlation test
Course subject(s)
Module 4. Data Collection & Analysis
Pearson’s r Correlation Coefficient
As we fully understand that your statistical knowledge may have gone a little rusty, let us briefly recap Pearson’s r Correlation test.
When you carry out a Pearson’s Correlation test you calculate Pearsons’s Correlation coefficient r which is standardized form of the covariance. This coefficient is a measure of relationship and can have a value between -1 and + 1, with -1 a perfect negative relationship and + 1 a perfect positive relationship. If Pearson’s r = 0, no relationship exists at all.
When to use?
You may only use Pearson’s r as a measure of correlation if your data is continuous and meets parametric assumptions. If your data does not meet parametric assumptions you could calculate Spearman’s rho as a non-parametric alternative.
For both tests, any relationship found is significant (as in meaningful) when it has at least a 95% confidentiality level, which means that the probability value, also known as p-value is smaller than .05.
Typically, values of r = ± .1 are seen as a small effect size, r = ±. 3 as a medium effect size, and r = ± .5 as a large effect size.
Please note: Correlation does not imply Causation!
For more information on statistical tests, we refer you to the statistics manuals in the Recommended Reading of this module.
![Creative Commons License](https://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png)
Efficient HVAC Systems by TU Delft OpenCourseWare is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Based on a work at https://online-learning.tudelft.nl/courses/efficient-hvac-systems//