CAMY claims this is "the first public analysis of underage traffic to alcohol Web sites, finding that nearly 700,000 in-depth visits to 55 alcohol Web sites during the last six months of 2003 were initiated by underage persons." The data comes from comScore Media Metrix, but there's no figuring out from the report just how many actual people under 21 were tracked visiting these sites. ("In-depth visits" are visits exceeding two page views, which presumably excludes visitors deterred by age verification pages.)
CAMY says 700,000 visits is around 13% of the traffic to the booze company sites that were analyzed.
Taking the 700,000 figure at face value, here's another (quick and rough) way to gage the magnitude of the supposed problem: There are around 65 million people in the U.S. ages 5 through 20. Assume 59 percent of them use the Internet (that's based on 2001 data for 5-17 year olds) and that each Internet user visits 1,000 non-unique Web sites per month (based on this). Over a six month period, that's: 65,000,000 x 0.59 x 1,000 x 6 = 230,000,000,000 total visits.
So 0.0003 percent of Web site visits by underage youth are to alcohol company sites, which CAMY describes as "attractive and appealing to underage youth." Does this worry you?