statistics
-
Bad idea, ASA
While I support the idea of giving the Bureau of the Census more autonomy and making it less susceptible to political influence, I really don’ t think that passing a bill during a lame duck session of Congress, especially after the historic change we saw in November, is any expression of the Will of the Read more
-
Friedman’s “Law of Statistics”
Kaiser Fung craps all over the latest Tom Friedman column, deservedly. Apparently Friedman never learned any quantitative critical thinking. Read more
-
I lack confidence in Confidence Intervals
From a comment I posted over at Will Briggs’ excellent site: Confidence intervals crack me up. When I teach these to my students I remind them that the word “confidence” has two opposite meanings, and that confidence intervals have much in common with confidence games. If you go the extra mile and explain that a Read more
-
Bayes, lingustics, and Alan Turing
Language Log has an interesting post on using Bayesian statistics to combine several items of weak evidence to arrive at a strong conclusion. Better yet, it contains an interesting reference to the cryptanalysis work of Alan Turing! Read more
-
Take two grains of salt and call me in the morning
Pertussis (whooping cough) is coming back, and it IS a serious disease. Bullshit analysis like this, committing a glaring ecological fallacy, doesn’t help. Unless ethnic group and socio-economic status is tied directly to each individual case of whooping cough, we don’t know–and cannot infer–jack. Tip from the Instapundit, who should have been a bit more Read more
-
Wear your farookin’ helmet!
San Antonio artist Chuck Ramirez has died from from a head injury in a bicycle accident; he was not wearing a helmet. Some folks don’t wear helmets because they think the helmet makes them look foolish. Does this guy look foolish? When I teach about odds ratios and contingency tables, I use the results of Read more
-
Better than pretty
Pretty R is a syntax formatter for R, it generates highlighted and properly indented R code suitable for including in an HTML document. Better yet, all the R keywords are linked back to an online reference manual at inside-r.org! Check it out. # multiple comparison with aov s <- read.table(“dogFoodSales.txt”,header=T) # fit the Read more
-
Problems in evidence-based medicine
Ponder this the next time someone in the reality-based community tells you the “science is settled” based on peer-reviewed studies. Tip from the Instapundit. Read more
-
Sentence first — verdict afterwards
Uh oh, I think they’re on to us. Tip from Will Briggs. Title from this. Update (15 October). Kaiser Fung reminds us twice about story time. Tip from Andrew Gelman. Read more
-
Logistic Regression for the People!
…with Bayesian priors no less. This is a neat application to help identify your risk factors for migraine headaches, and it’s FREE! Tip from Andrew Gelman. Read more
The Random Texan 