Statistics

Statistics Mailbag

Because of my new gig and then my two-week sojourn at Cornell I am way behind in acknowledging all the emails and blog tips I have received. Nearly all stories here originate as a link from readers, and for these I am very grateful. Thank you to all who send these in and apologies for not writing back to everybody personally.

Since I have no hope of writing a separate post on all these juicy tips, here today, in no order, are some of the older ones which deserve a wide audience. I might revisit some of these later and expand on them.

General Bad Statistics


Your Brain Scan Looks Different on Mac and PC

A team…took data from 30 brain scans and analyzed them using a package called FreeSurfer…to measure the size of different parts of the brain.

They ran the software on PCs, and also on Macs running different versions of Mac OS, each time using the software to measure the size and thickness of various structures of the brain…Across most sections there was at least a 2-5 percent variation in the answers.[other] answers diverged by as much as 15 percent.

Counties that enjoyed better weather on tax day had more people sign up to become Tea Party organizers. And the paper.

What’s more, the Tea Party experiment shows that the activism catalyzed by those sunny days translates into real political influence. Politicians whose districts were sunny on tax day voted in a more reliably conservative fashion throughout 2009 and 2010. Indeed, the absence of rain in a congressional district on April 15, 2009, made its representative 8.7 percentage points more likely to vote against the Affordable Care Act. Had the weather at those early rallies been sunnier, it’s possible that Obama’s signature legislation wouldn’t have passed.

Why Women Choose Bad Boys

Women choose bad boys because their hormones make them…

Doggy daydreams: brain scans reveal Fido’s thoughts

Now we can really begin to understand what dogs are thinking. We hope this opens a whole new door into canine cognition, social cognition of other species.

Climatological Bad Statistics

Climate Change Denial in the Classroom (pdf)

models suggest…

Climate sensitivity to the lower stratospheric ozone variations Looks like another smoothing before regression “study.” A naughty no-no and guarantor of over-certainty.

Doctored Data, Not U.S. Temperatures, Set a Record This Year

To most people, the hottest temperatures ever “recorded” would imply that quality controlled thermometers registered higher readings during the past year than had ever occurred before. If you believe that this is what the National Oceanic and Atmospheric Administration (NOAA) means by hottest temperatures ever “recorded,” then you are wrong.

Misc

Email from John Cook to which we can reply Amen, brother.

Matt,

I’ve started reading The Discarded Image by C. S. Lewis. Here’s a quote from it that sounded to me like something that might appear on your blog:

“In our age I think it would be fair to say that the ease with which a scientific theory assumes the dignity and rigidity of fact varies inversely with the individual’s scientific education.”

— John

Categories: Statistics

5 replies »

  1. My life-long period of observation leads me to believe the reason why most women choose the “bad boy” is to stick it too their mothers.

    And has the data supporting “Doggy Daydreams” mean run on PCs and Macs and compared? Don’t mean to hound you about this but some degree of consistency is desired.

  2. What’s all this five-sigma stuff I’m hearing about a Briggs boson?

  3. What do you think of Hansen’s “climate dice” paper, in which he tries to predict heat waves from distributions of temperature anomalies? Besides other things, isn’t this really trying to extract something meaningful from an abstraction of an abstraction?

Leave a Reply

Your email address will not be published. Required fields are marked *