ABC/WaPo-National: Trump still leads

[Donerail]

Link

Trump 24
Walker 13
Bush 12
Huck 8
Rubio 7
Paul 6

Poll done Thursday-Sunday, with Trump's support "significantly weaker" (single-digits) among the small group sampled Sunday.

[King]

#prayfortrump

[MT Treasurer]

Dems:

Clinton: 63%
Sanders: 14%
Biden: 12% (most of his support goes to Clinton; with Biden out, she gets 68%)

[Mehmentum]

Trump literally doubling Bush.  Wow.  We'll see just how far he falls.

[Skye]

I'd rather wait until the fallout of those McCain comments takes their toll... or not.

[Simfan34]

Jesus. Hopefully the Sunday group is not just a fluke.

[Mehmentum]

Jesus. Hopefully the Sunday group is not just a fluke.

Yeah, another poll said there was no difference after the comments.  We'll just have to wait and see.

[Obama-Biden Democrat]

Damn, this possible Trump gaffe worries me. It is a stupid move to attack a veteran's service in a extremely jingoistic war hawk murrica 1st party like the Republicans.

[RJEvans]

Trump is done. He is in the single digits not 24%. As a Democrat, I'm a little disappointed. These comments won't go over well with primary voters.

[Likely Voter]

when you look at the rise of Trump it has come at the expense of pretty much everyone except Bush (and maybe Walker). If Trump collapses will his support get spread around or go disproportionately to one or another. Cruz seems to be betting he is going to get it. As Claire McCaskill tweeted today “There’s something unseemly about Cruz following Trump around like a lost puppy, hoping to get his leftovers when he finally flames out,”

[Mr. Morden]

Dems

Clinton 63%
Sanders 14%
Biden 12%
Webb 2%
O'Malley 1%
Chafee 0%

GOP

Trump 24%
Walker 13%
Bush 12%
Huckabee 8%
Rubio 7%
Carson 6%
Paul 6%
Cruz 4%
Perry 4%
Christie 3%
Jindal 2%
Kasich 2%
Pataki 1%
Santorum 1%
Fiorina 0%
Graham 0%

[Lief 🗽]

The Sunday group would have been a sub sample of like 50 people. You can't really draw any conclusions from that.

[Simfan34]

The Sunday group would have been a sub sample of like 50 people. You can't really draw any conclusions from that.

They said that it was statistically significant.

[Beezer]

How can a sample size of 50 produce anything "statistically significant"?

[Brittain33]

How can a sample size of 50 produce anything "statistically significant"?

Margin of error for a sample size of 50 is 14%. If his number was like 5% on Sunday, that would be a statistically significant drop from 24% even counting the MOE on the larger non-Sunday sample, too.

[Donerail]

How can a sample size of 50 produce anything "statistically significant"?

Margin of error for a sample size of 50 is 14%. If his number was like 5% on Sunday, that would be a statistically significant drop from 24% even counting the MOE on the larger non-Sunday sample, too.

I believe I saw that his support on Sunday dropped into single digits.

[Bakersfield Uber Alles]

How can a sample size of 50 produce anything "statistically significant"?

Margin of error for a sample size of 50 is 14%. If his number was like 5% on Sunday, that would be a statistically significant drop from 24% even counting the MOE on the larger non-Sunday sample, too.

I believe I saw that his support on Sunday dropped into single digits.

I saw this as well. Not an exact number though.