Using two different y axis scales is really not good datavis. Can cause a bunch of misleading conclusions.
Edit: not that I’m saying the correlation in the graph is unfounded. I’m just saying it’s better to adhere to good datavis practices so you don’t open yourself up to this type of criticism
Using two different y axis scales is really not good datavis. Can cause a bunch of misleading conclusions.
Edit: not that I’m saying the correlation in the graph is unfounded. I’m just saying it’s better to adhere to good datavis practices so you don’t open yourself up to this type of criticism
https://blog.datawrapper.de/dualaxis/
At least color the Y axes the same as the data they’re representing