Weird correlation between Maine’s divorce rate and margarine consumption

You’ve probably heard this before: Correlation does not demonstrate causation.

The point of the statement is that, when two variables tend to go up or down at the same time, they can have absolutely nothing to do with each other. In statistics talk, when two variables are highly correlated and there isn’t a cause and effect connection (also called no causal mechanism), that’s a spurious correlation.

OK, enough with the lesson. Here’s what you came for, a chart showing the correlation between Maine’s divorce rate and margarine consumption.

From: Dylan Matthews, "The best illustration you'll see that correlation doesn't equal causation," (May 13, 2014).

From: Dylan Matthews, “The best illustration you’ll see that correlation doesn’t equal causation,” (May 13, 2014).

As you can see at the bottom right of the chart, these two variables have a whopping high correlation of .992558, or just 0.007442 away from a perfect correlation between two variables — 1.0.

Margarine l Creative Commons

Margarine l Creative Commons

There’s no reason why lower margarine consumption over time would lead to fewer divorces — or vice versa — so, yes, this is a spurious correlation!

Check out the source of this chart and others at See how the number of lawyers in California correlates with how much money is spent on pets in the U.S. – and more!

Amy Fried

About Amy Fried

Amy Fried loves Maine's sense of community and the wonderful mix of culture and outdoor recreation. She loves politics in three ways: as an analytical political scientist, a devoted political junkie and a citizen who believes politics matters for people's lives. Fried is Professor of Political Science at the University of Maine. Her views do not reflect those of her employer or any group to which she belongs.