Correlation report

The Correlation Report applet allows users to measure how closely survey fields are related (or "correlate"). Users may select two or more variables to analyze.

Types of correlation

When creating scatter plots of how variables relate to each other, the correlation of these variables can fall into one of three different categories: positive correlation, negative correlation, and no correlation. Correlation values are measured by the correlation coefficient r, also known as "rho." Correlation values range from -1 to 1; the closer the r-value is to ±1, the stronger/more likely the correlation is between the variables. Variables that have correlation values close to 0 indicate little to no correlation between the values (e.g., height and IQ should have no correlation).

• Positive correlation: As one variable increases in value, so does the other. The more positively correlated two variables are, the closer their r value will be to +1. An example of this would be height and shoe size - as one's height increases, often their shoe size does as well.
• Negative correlation: As one variable increases in value, the other decreases in value. The more negatively correlated two variables are, the closer their r value will be to -1. An example of this would be time spent studying and time spent on video games - as one's time spent studying increases, their time spend playing video games should decrease, and vice versa.
• No correlation: There is no apparent relationship between the variables, with an r value close to or equally 0...

Creating a Correlation Report

To access the Correlation Report applet, select Analytics > Correlation Report from the Survey Navigation menu.