Selecting a Statistical Test
Pearson Product Moment Correlation Coefficient
Research Design
The Pearson's r is used when you need to conduct a correlational study. A correlational study is a quantitative method to determine the degree to which two variables from the same group of subjects are related. The two variables (x, y) cannot be manipulated. This test does not take into consideration which variable is used as dependent or independent, because the correlation coefficient r will be the same.
Levels of Measurement
In order to use this statistical test, the data should be interval or ratio. The variables could be both interval, both ratio, or a combination of interval and ratio.
The Pearson Product Moment Correlation Coefficient (PPMCC) is also know as "Pearson's r", "r", or "Pearson's Correlation Coefficient".
The Pearson r is intended to determine the correlation of 2 variables from the same group of subjects.
Number of Groups
Sample Type
N/A.
Description of Test
The Pearson Product Moment Correlation Coefficient is a statistical test used to measure the degree to which two variables relate to each other. If they are correlated, the change in variable one is accompanied with change in variable two, either in the same or opposite ways.
In the results, r can take values from -1 to +1, where r>0 indicates a positive relationship and r<0 indicates a negative relationship. This is, the most closer the coefficient is to +1, the strongest the relationship of the variables is in the same direction, and the most closer the coefficient is to -1, the strongest the relationship of the variables is in opposite direction. To better understand how close the relationship is, the following scatterplots indicate the strength and direction of the relationship based on how close data is to the line:
Once we calculate the value of r, the following table offers a guideline to determine the strength of the relationship:
Examples of Research Questions That Can Be Answered With This Test
-
Is there a statistically significant relationship between grades, as measured with GPA, and time spent in the library, as measured in hours?
-
Is there a statistically significant relationship between the number of calories ingested daily and hours of sleep?
Notes
-
To conduct a Pearson correlation, both variables should be normally distributed.
-
This test is sensitive to skewed distributions and outliers. If the conditions are not met, then use Spearman's correlation.
-
If you have Ordinal data, then you can use Spearman's Correlation Coefficient.
-
If you need to find a correlation using more than two variables, then you can use the Multilinear Correlation.
-
Correlation does not determine causation.
Video Explanation
Additional Resources
For more information about this test, please visit the following links:
