Jul 31, 2013
The causes — or not — of home sales (or not)
31 Jul 2013
Posted by Andrew Kantor
Here’s 30 seconds on causation vs. correlation. It’s important.
Just because one thing happens around the same time as another thing, it doesn’t mean they’re connected. It’s a mistake a lot of people make.
Here’s a simple and clear example: A new principal comes to a school. A year after she arrives, reading scores are up eight percent. It must be because of her, right?
That’s correlation. The two things happened around the same time.
But that year the school district’s lines were redrawn, and it had rained a lot more than usual; less outdoor recess = more library time.
So maybe it was the principal’s new policies that increased scores, or maybe it was the change of student body, or maybe it was the extra time students had in the library. We don’t know the actual causation.
Keep this in mind when you see headlines like "Pending Home Sales Squeezed by Changing Mortgage Rates." (That happens to be from Realty Times.) Home sales may have slipped from month to month just as interest rates rose slightly, but that doesn’t mean one caused the other.
A lot of other things also happened. We had some crazy-hot weather, for example, so why not "US Heat Wave Keeps Home Shoppers Away"? Or "Birth of Britain’s Prince George Has Americans Glued to TV Instead of House Hunting"?
Those things also occurred around the same time, and they make about as much sense as a family saying, "We were gonna buy a house, but mortgage rates went up two-tenths of a percent, so forget it."
In the February/March issue of Commonwealth we wrote about how to read the real estate numbers and news — about which numbers actually mean things, and how it’s easy to spin them. If you haven’t read the piece, you should.
There’s nothing wrong with conclusions, but you might not want to jump there too quickly.