I really like this blog post.

Here is how I stumbled across it. When I suggested that correlations being caused by a third factor needed a snappier, catchier name in "Cousin Causality," Chris Chabris tweeted that it was called "third-variable causation" in psychology. Though neither this nor the other term in use, "shared causation", meet my standards for being snappy and catchy, knowing the phrase "third-variable causation" led me to a search on Twitter that turned up Julia Rohrer's post.