Here we have iPod sales and housing starts, two metrics so thoroughly unrelated that their 95% negative correlation seems less like a discovery and more like evidence that the universe is playing a joke at our expense. As Americans bought fewer white earbuds, they bought fewer houses, which is roughly equivalent to finding that umbrella sales correlate perfectly with the divorce rate in Luxembourg. The correlation is so strong it feels almost intentional, as if someone in 2005 had decided that every iPod sold would require the cancellation of one foundation pour somewhere in Ohio.
What's actually happening is far more mundane than cosmic mischief, which is both disappointing and oddly reassuring. The 2008 financial crisis acts as a sort of economic guillotine falling through both datasets simultaneously—housing starts collapsed from about 2.1 million annually in 2006 to roughly 554,000 by 2009, while simultaneously people with evaporating home equity and frozen credit decided that yes, actually, they could live without a fourth-generation iPod. Both trends also reflect the general shift from physical accumulation (houses, gadgets) to digital services (streaming, renting), though that's more visible in hindsight than in the raw numbers. The real story is just two separate casualties of the same economic earthquake.
And so we arrive at what Spurious does best: the humbling reminder that coincidence and causation are closer cousins than we'd like to believe. We are pattern-recognition machines operating in a universe of overwhelming coincidence, which means we will always, always find what we're looking for. The iPod and the housing market never knew each other at all.
As an Amazon Associate, getspurious.com earns from qualifying purchases. Learn more.
Want to learn more about why correlations like “iPod units sold” vs “New housing construction starts” don't prove causation? Read our guide to statistical thinking.