2016 Postmortem
Related: About this forumWhy do Clinton and Trump do better in robo-call polls and worse in live cell/landline phone polling?
This is a documentable phenomenon.
Here is the New Hampshire polling according to ALL the robo-call polls considered by either RCP or Huffington Post's Pollster:
Here is the New Hampshire polling according to ALL polling considered by either RCP or Huffington Post's Pollster except with the robo-call polls excluded:
The inclusion of robo-call polling in the pollster aggregation models makes the New Hampshire race seem like a closer race than the impression you would draw from all the other polling. The phenomenon also holds true in Iowa, where the robo-call polls make Clinton's lead seem bigger than other polling indicates. This same phenomenon also favors Trump in Iowa and New Hampshire in the similar manner that it favors Clinton.
Any thoughts about the cause of this phenomenon?
RandySF
(58,823 posts)Respondents feel less shy about stating their true choices.
Attorney in Texas
(3,373 posts)admit the preference in a live phone poll so that might explain the difference (538 discussed possibilities but came to no conclusions).
I think that is not likely because while that might make some sense with Trump (people feeling embarrassed to tell a live interviewer that they support someone who has so many trollish policies) but that does not seem to apply to Clinton (I'm a Sanders supporter but I cannot see why anyone would be embarrassed to support Clinton).
RandySF
(58,823 posts)But I've reached the point where I don't take the time to answer surveys (we get a lot), but I'm more likely to push a button.
Attorney in Texas
(3,373 posts)RandySF
(58,823 posts)Renew Deal
(81,859 posts)Basically educated people won't admit to supporting him. So the theory is that his polls are stronger than the numbers show. Could this be bad for him in IA?
http://www.democraticunderground.com/10027491134