Also, yes … I was disappointed that they made the claims about the special seat-belts without commissioning anything on them. What they’ve done is highlight a potential problem, but their “solution” is wholly untested. Granted, I may be a bit biased here, but if you’ve not tested your solution, you can’t claim much of anything about it.

[I'm also a bit miffed at the test engineer who didn't want to test anything. What the hell? Most test engineers I know live to break stuff and bust up people's hypotheses.]

For children younger than roughly 24 months, seat belts plainly won’t do. For them, a car seat represents the best practical way to ride securely, and it is certainly an improvement over the days of riding shotgun on mom’s lap.

Also, as Edmunds pointed out, Levitt’s and Dubner’s data hasn’t gone through peer review. I find their analysis interesting but not enough to draw a conclusion from. I am bemused that their latest response was entitled, “We Are Not the Only Ones Who Think Child Car Seats Don’t Work Well” whereas their earlier claims boiled down to “specially modified seat belts (which do not currently exist) would work as well as car seats.”