To read all comments associated with this story, please click here.
Hmm, I don't know if I agree with that. In Sweden where (some) banks are in league with the government and responsible for issuing electronic "identity cards" I would argue that statistics from polls where the participants sign with their "electronic identity card" is reliable, given that all other parameters are identical (or adjusted) to match a "real world poll". The only major source of error would be that a certain group of people tend to acquire an electronic "id". However given Swedish GINI and literacy coefficients that would not be very relevant.
-----
As an explanatory note everyone in Sweden (from birth) has a "personnummer" (think "social security number" in US) that follows the person through his/her whole life. Opening a bank account requires valid identification complete with "personnummer". Getting a electronic "identity card" requires an bank account.




Member since:
2006-02-10
I think the crux of your article was that online polls are so unreliable they are basically garbage. So, if you really want to know what people think (or want), I suggest you trust an experienced statistician more than a web designer!
Everytime I hear about an "online poll" I know it's either a misguided attempt or plain propaganda. Real statistics are hard and obsessed about little details.
The other part, about some people being more influential than they would deserve otherwise due to their popularity... I think this is the way things have been for better or worse since antiquity, at least in (semi)free societies.