Mark links to a presentation by Steven Levitt (of Freakonomics fame) that claims that the effectiveness of child car seats is dubious--that more or less, death rates for children (not infants) are about the same whether they're in their safety seats or not.
Now I've got my qualms about uncritically accepting Levitt's work; he is, after all, the one responsible for the false claim that abortion lowered crime in the 1990s. In the video, he also neglects as "unimportant" the units that are used to characterize the performance of child seats--as an engineer, I know better.
That said, I must give him credit for using what data he had to raise the question of whether child safety seats actually save lives. I dare suggest that it's time with this issue, as well as with the broader issue of seatbelt usage, to start asking that those who would buckle us up in ever more complicated harnesses actually provide the accelerometer data and hypotheses tested before we accept new seatbelt and child seat laws.
In Which I Condemn Metrosexuality - I am -again- reheating up a leftover post from a time when I wrote from better-formed thoughts. This one is from 5 years ago as well, with a couple of edit...
4 hours ago