[evldave]

Originally Posted by Michael1

If you read the post a little more carefully before shooting off at the keyboard, you'll notice the data was normalized per million miles.

Michael

No, it's actually per 100 million miles. Just checked the NHTSA website.

http://www-nrd.nhtsa.dot.gov/pdf/nrd...004/809908.pdf

AND, motorcycle accidents actually DROPPED when DRLs started being used (95-98).

1994 22.66
1995 22.73
1996 21.78
1997 20.99
1998 22.31
1999 23.46
2000 27.67
2001 33.17
2002 34.23
2003 38.93

Besides, none of this matters. I'm guessing that there are so many statistical variables, there there's no way with any level of confidence (at least a 95% confidence interval) to prove anything one way or the other. A common statistical fallacy is to assume since the average number changes that it's the result of something - without a real statistical analysis, there's no way to tell.

An example of a real statistical analysis is this:

http://www.nhtsa.gov/portal/site/nhtsa/template.MAXIMIZE/menuitem.51ea2eb4d278d13bc22cf37490008a0c/?javax.portlet.tpst=3c0dd0fb9371f21ab25f5ed01891ef 9a_ws_MX&javax.portlet.prp_3c0dd0fb9371f21ab25f5ed 01891ef9a_viewID=detail_view&javax.portlet.begCach eTok=token&javax.portlet.endCacheTok=token&itemID= 8c184e5e1adaff00VgnVCM1000002c567798RCRD&viewType= standard