Environmental and Science Education
Mathematics Education
by Edward Hessler
It seems that everyday we see a new headline suggesting a relationship between some X and some Y as though it was causal.
The political season is rich in such reports, e.g., relating one's politics/political leanings-Republican, Independent, Green Party, Democrat, liberal, conservative, with personality.
Maria Konnikova's wrote a short piece in The New Yorker on the so-called relationship between politics and personality. She reviewed research done by Brad Verhulst, Virginia Commonwealth University and also talked with him about it. This may be one of the reasons that the subheading for her essay reads "most of what you read is malarkey." There may be relationships but they are all too often stated much too strongly.
In this essay Konnikova mentioned Tyler Vygen's website, Spurious Correlations. This is the main purpose of this post: to point you to his web site.
Vygen, a Harvard law student, has a love for numbers and science. He is not a math/stats person. About the charts on his website he writes that they "aren't meant to imply causation nor are they meant to create a distrust for research or correlation data. Rather, I hope the project fosters interest in statistics and numerical research." (emphasis added)
Vygen's graphs are free for the taking. Anything he posts is "released under a Creative Commons Attribution License." Such a deal!
Konnikova's essay is a lesson on how one comes to trust research by first asking some questions about the data.
No comments:
Post a Comment