Interpret Your Correlation

Find relationships in data with correlations

Jennifer Funk avatar
Written by Jennifer Funk
Updated over a week ago

Find the relationship between two sets of data by using the 'Correlations' feature. For example: Ever wonder if there is a relation between residents using food stamps and their educational attainment? If so, how much? Is the relation stronger or weaker than you expected? Now, you can discover what data are related, support your findings, and determine how strong they are.

  1. In the right hand corner of your map, you'll notice a white box that shows the strength of the relationship between the two sets of data you have selected (i.e.: Correlation: + 0.09 ⍊ 0.12)

  2. Interpret your correlation: you'll notice that there will be a text description of the correlation, the correlation coefficient value, and the margin of error.

The text description of the correlations are paired with the following correlation coefficient values.

No Correlation: 0.3 to -0.3
Weak Correlation: 0.5 to 0.3 and -0.3 to -0.5
Moderate Correlation: 0.8 to 0.5 and -0.5 to -0.8
Strong Correlation: 1 to 0.8 and -0.8 to -1

If your correlation is labeled invalid, that is because the margin of error is too high. The margin of error is listed after the correlation coefficient. 

If you get unable to correlate, it means you have fewer than 20 features (geographies) in your map view. Since correlation is about the relationship of values from both datasets, you need a minimum of 20 geographies visible to calculate a correlation.

Still not sure how to correlate your data? Watch an example here:

Did this answer your question?