OH-Quinnipiac: Christie/Clinton tie; Clinton leads Paul; Biden loses to GOP

[Mr. Morden]

Quinnipiac poll of Ohio:

http://www.quinnipiac.edu/institutes-and-centers/polling-institute/ohio/release-detail?ReleaseID=1914

Christie 42%
Clinton 42%

Clinton 47%
Paul 44%

Christie 50%
Biden 32%

Paul 49%
Biden 40%

[Mr. Morden]

fav/unfav among all voters:

Christie 48/16% for +32%
Paul 43/26% for +17%
Clinton 52/44% for +8%
Biden 41/48% for -7%

Christie has broad support across party lines, while opinions of Clinton are more polarized:

Christie fav/unfav among…
Dems: 37/19% for +18%
GOP: 55/14% for +41%
Indies: 53/14% for +39%

Clinton fav/unfav among…
Dems: 86/9% for +77%
GOP: 18/80% for -62%
Indies: 55/43% for +12%

But one thing that may be inflating the fav/unfav differential for Christie and Paul is that Dems are less likely to know who they are.

% saying they don't know enough about Christie to have an opinion:
Dems: 43%
GOP: 29%

% saying they don't know enough about Paul to have an opinion:
Dems: 38%
GOP: 22%

[Mr. Morden]

The age crosstabs continue to be interesting in the 2016 polling.  In Christie vs. Clinton, Clinton does better among the young than the old, though the difference isn't very large compared to recent elections:

Christie vs. Clinton:
18-29: Clinton +6
30-44: Clinton +9
45-64: Christie +3
65 and older: Christie +5

But in Paul vs. Clinton, Paul actually wins the young while losing every other age group:

Paul vs. Clinton:
18-29: Paul +7
30-44: Clinton +14
45-64: Clinton +1
65 and older: Clinton +5

In both cases, Clinton's best age group is 30-44 year olds, even though most of the PPP polling shows that that's her worst demographic!

[Mr. Morden]

Are they using a likely voter model at all?  The survey writeup talked about a sample of "registered voters".  Aren't they just calling people up, weighting by demographics (but not party ID) to get a registered voter sample, and then they get what they get?  I don't think they make any assumptions about party ID per se.