Monday, March 25, 2013

Tuesday, March 12, 2013

Jessie Spano's Caffeine Intake

In honor of Daylight Savings Time and my own caffeine addiction.

This will always be the funniest graph ever, p < .05.

Property of Nathaniel James

Monday, March 11, 2013

PHD Comics, 1/20/2010

Jorge Cham of PhD Comics quickly summarizes the problems that can occur as research is translated to the masses via the media. Funny AND touches on CI, sampling, psychometrics, etc.

Property of phdcomics.com

Thursday, March 7, 2013

Brett Keller's "An incredilby detailed super statistical Hunger Games survival analysis" via io9.com

Sometimes I think I'm a big nerd. Then I read about an even bigger nerd and I feel better about myself because I'm less nerdy but then worse about myself because I'm not as hard core as I once thought I was. Especially when their nerd focus is on 1)statistics and 2) The Hunger Games.


Monday, March 4, 2013

Newsweek's "About 40 percent of American women have had abortions: The math behind the stats"

This exercise encourages students to think critically about the statistics that they encounter in the media.  Note: This data is about abortion. When I use this exercise, I stress to my students that the exercise isn't about being pro-choice or anti-abortion, it is about being anti-bad statistics.

Writer Sarah Kliff published an article in Newsweek about de-stigmatizing abortion. In the article, she makes the claim that 40% of American women have had abortions. Some readers questioned this estimate. This is her reply to those readers.

I challenge my students to find flaws or possible flaws/points of concern in the mathematics behind the 40% estimate. Some points that they come up with: a) The data doesn't include minors, b) the data doesn't include women who were alive between 1973 (Row v. Wade) and 2004 and but died/moved out of the US before the 2005 census data (which was used in her calculations), c) data estimates that about half of women having an abortion at any given time have had a previous abortion, d) the author has is pro-choice, which may or may not be an issue, and e) her data for total number of legal abortions comes from a pro-choice organization, which may or may not be an issue.

Again, my example isn't about pro-choice/anti-abortion, it is about judging the veracity of data being used for persuasion.

Final, in order to equally inflame my anti-abortion students, I like to follow this exercise up with this clip from The Daily Show with Jon Stewart in which Stewart rips apart anti-abortion (now former) Senator John Kyl for making up statistics in order to attack Planned Parenthood.