This is a new poll from the Public Religion Research Institute (PRRI). They are a nonprofit, nonpartisan research and education organization. Their business is polling. They conduct public opinion polls for clients like the Brookings Institute, and have done a number of surveys tracking influential sub-groups and issues in our country like: tea party voters, immigration, millennials, transgender rights etc. All this is to say that these guys are legit, and they seem to have their finger on the pulse of America because it is their business to do so. They are probably more legit than our friends at Gravis, Rasmussen, Q, UPI/CVoter, Morning Consult, USC/LAT and the other trackers which dominate so much of the polling averages and polling discussion. All this is as a preface to say that their polling result should be taken no less seriously than those other polls, but they are probably deserving of a standing closer to Pew (Pew Jr if you will). Here are the results:
Dates: August 10-16 2016 (during week 1 and 2 of the Olympics)
Sample Size: 1630 RVs, +/2.9 MOE
Top Line Result: Clinton 48%, Trump 35%, Johnson 2%, Other 2%, Refused/DK 10%
Subgroup Result:
*Trump leads among white voters overall 44-39.
*Clinton leads 67-18 among latinos and 85-15 among AA voters.
*Trump leads 50-32 among non college educated white voters.
*Clinton leads 51-33 among college educated white voters.
*Clinton leads among white catholic voters 44-41. Clinton is +8 among white women and -19 among white men (better than Obama)
Some Thoughts:
Numbers are obviously #greatforher. The survey was completed about 1 week ago, so some may find it dated, though I don’t think much has really changed in this dead period of the race despite the MSM efforts to pump up Trump. It overlapped to some degree with the Pew Poll, so consider this as the high end and Pew as the low end, with the average falling around 6-8 points. Pew had higher numbers for Trump among latinos and young people that conflict with other reputable polls (which is why their number was lower).
I think the subgroup trends are very consistent with what we’ve seen in reputable public polling. Trump has a strong lead among non-college educated whites, but he is not himself getting a high enough percentage (though his margin is good) to overcome his losses among college educated whites. In this subgroup sample, Trump doesn’t do any better than Romney did (good for Team Blue). This is why Clinton is well ahead in VA and CO and other states and why the race is still nominally tighter in IA and NV. The variance is whether Clinton is pushing into the low 30s and whether Trump is exceeding 60%. This is why Clinton is going after outsourcing in its latest ad. The difference between having to fight it out in a state and having the luxury of moving resources to new battlegrounds is to keep Trump’s numbers among non-college educated whites at 60% or below.
www.prri.org/...