Correlation Heatmap

Welcome to the Correlation Heatmap. This custom tool allows you to interact and engage with the data. You have control. Use it to choose the variables that matter to you and build your own data charts. Share them with colleagues and friends, or print them out for further use. Of course, remember to download the Cities of Opportunity report in its entirety, or select from the interviews, policy analysis, or indicator discussions.

The tool allows you to correlate the similarities and dissimilarities in the comovement of different variables—essentially tracking the tendency of two variables to move together or apart—revealing interesting and potentially causal relationships, notable among them shedding light on families of urban economies and how they behave.

Correlation analysis tool

Use the tool below to get to the heart of the study.

  • Reset heatmap
  • Print friendly
 
 
 
 
 

 
 

What this data means

Correlation analyses map the covariation or comovement of two statistical variables, commonly using hot or cold colors to code the direction. Positive correlations occur when an increase in X is associated with an increase in Y, or, the variables move in a similar track. In negative correlations, X rises as Y falls, or the tracks move opposite each other. In this chart at left, darker red indicates stronger positive correlation. Darker green shows stronger negative correlation.

Causal connections are more likely to be occurring at the extremes of positive and negative correlation, the closer that covariation gets to +100 percent or -100 percent. (The same variable would be +100 percent positively correlated with itself.)

Correlation charts, also known as heat maps, depict in the upper left an average correlations row and column followed by a headline row or column. The average provides a benchmark for comparing how high or low a correlation is to others. The headline highlights what the particular analysis is showing. (For instance, the city economy heat map includes all 23 variables that go into creating a city’s economic profile.) The headline also serves to sort the entire matrix because each row and column is reordered from raw input to reflect the relative strength of the correlation with the headline indicator.