i’ve been sitting on this draft post for about a week thinking that i was going to return and add more to it. usually the best idea is to just publish as they get written.
there is an excellent reposting over on boing boing about the fabrication of news and regurgitation of press releases in the mainstream media (and everywhere else as a result). the article focuses specifically on the yearly exercise that surrounds reporting consumer holiday spending, and does a great job of it.
on the same front, over at the big picture barry ritholtz is keeping his eye on the way statistics are being abused and reported as fact. he posted about the new home sales numbers released this week, and this little piece made me actually laugh aloud:
Sales of new one-family houses in October of 2005 were at a seasonally adjusted annual rate of 1,424,000 . . . This is 13% (+/- 17.7%)* above the revised September rate of 1,260,000 and 9.0% (+/-18.2%)* above the October estimate.
look at that margin of error. not only are the sales numbers rendered completely devoid of meaning, i wonder how they were able to pin down a margin of error with that degree of specificity. +/-18.2%? where did that two tenths of a percent come from? my gut tells me they pulled all those fiigures out of their asses. (yes, that is an opinion, and i am certain that a large number of people with advanced degrees were paid lots of money to, emm, extract those figures). anyhow, barry has created a new catagory on his blog called “data analysis” that i will be following.
i guess the real question that i have through all of these individual incidents is…why aren’t journalists digging into these stories more? why not point out that data or the press releases in your email inbox don’t make sense? or that they make complete sense based on the bias of the organizations sending them?