Confusing. I would expect internet connectivity to go down as userbase expands making the console with the most sales the most likely to have the lowest internet connectivity... due to the audience being broader and less specialized.

Guess it shows the Wii generally has a broader enthusiast market then people think.

out of the 5,000 who played online 1000 had ps3s. we have no clue what the 20,000 pollers had

1000 play PS3s online.  Not had PS3s.

I have a PS3.  I haven't ever played it online.

While I have a PC and have played it online.

That isn't what the article says. 

"as last year 90 percent of online gamers were playing on the PC"

"last year 18 percent of online gamers played on the Wii"

etc.

It doesn't say 90% of PC owners gamed online last year, or that 18% of Wii owners did.  It says that out of the people who play online, that percentage was using that platform.

If only 29% of Wii gamers play online, what does that say of World at War?
There are 500k accounts and has sold close to a million. That would mean the online penetration would be about 50%.

Yeah noticed that.  The % in the OP messed me up when I was reading the article.  Still sample bias is unlikely in a group of 5,000.

The percentages should be fairly telling... and generally what matters... the "attach ratio" of online % would just mean something the other way as it would make things even worse then they look for the PS3 in this regard.  Rather then just being bad the numbers would disasterous.

^ well... considering i was talking about the 5,000 who played online.

if the market share represents the sample size which is logical then 9400 polled owned wiis. that would mean 15% of them play online. 7200 had 360s and 35% play online and 3400 ps3 owners with 29%

see my point:

360: 50%  with an pay service?

PS3: 20%  with an free service? 

Well, are you able to make multiple online accounts on the same Wii? If so, your logic doesn't work.

Not that I know of.