re: #86 The Mongoose

I do understand that. The DRI in this poll is an indication that there was a non-random bias in this particular poll.

Do you understand their previous polls have found basically the same split?

When I say it’s wrong, I mean it’s not taking an accurate random snapshot of the electorate. This may have just been the 1 time in 20 (assuming 95% confidence intervals) but in my experience it’s more likely to be because there was an undetected systemic bias in the “random” sampling methodology, or the weighting.

Then please demonstrate that, since they’ve laid their methedology out there for you to review.

Of course none of this is to say the President won’t win by 8…just that there is mathematical reason to have fairly high confidence his margin of victory would be outside the MOE of this Pew poll if the election were held today.

I think so too. I think this poll is somewhat of an outlier. I don’t get why you’re trying to use words like ‘wrong’ and, especially, I don’t get why you think that when doing a poll that includes a question about party affiliation, you think they should afterwards alter that number on party affiliation to reflect the ‘true’ number. I don’t think you get that’s one of the things being polled.