3 ms·

I'm not sure I understand what objections would be to the study you cite.

• andrewla 7y ago
  A telephone survey of 982 people called on landlines. How many total calls were made -- what was the response rate, and what has been done to de-bias any correlation between response rate and the outcome? And a self-reporting survey, over the phone, about government photo ids, without any mention being made of how they convinced people that the survey was genuine? And then there's just the standard phone survey stuff, where you up-multiply based on (once again, self-reported) attributes to match population demographics, causing everything to be wildly out of whack. This level of obfuscation of methodology is pretty high even by the standards of regular polling.