is released (the entire poll). Maddow stated it is not known if Sanders was omitted from the questions or the polling company did not release the results. She just got a peek preview.

While they actually represent 17% of The population?

That pretty much discredits the general election matchup results in this poll, an actual representation would have much closer results.

The results make since as far as the Latino only results, of course the numbers that matter will be swing states like always.

Where does it state that?

"MSNBC survey of 1,115 registered voters, including 432 registered Latino voters"

I just did it on the calculator, it's actually 39% which is still weighting the poll more than double actual Latino demographic representation.

All other polls that are more in line demographically have Clinton-Trump even with MOE the last couple of weeks for example.

Latinos were only 10% of the electorate in 012:

http://www.ropercenter.uconn.edu/polls/us-elections/how-groups-voted/how-groups-voted-2012/

so maybe with population growth and outreach it can be grown to 11%, 12% ???

I doubt any pollster would be dumb enough to project they will be 39% of the voting population. It could be the pollster oversampled them to get a better view of how they are voting but only used them as 10% -12% of the sample. I have seen African Americans oversampled the same way.

I can't believe any pollster would be that stupid, especially Marist University.

You need to know


% White
% African American
% Asian
% Other

Too much effort to fully unskew this poll with everything.

Politics usa and u guys are reading poll article incorrectly

From msnbc

The poll of hypothetical match-ups for 2016 was
conducted Aug. 26 to Sept. 9 of 1,115 registered voters. It has a margin of error of +/- 2.9 percentage points. The second match up polled 432 registered Latino voters. It has a margin of error of +/- 6.2 percentage points.

*****

Clearly, the 1115 registered voters refers to all voters. Also, just from a common sense stand point why would marist waste time and money on a poll with 40% latino that would end up being completely meaningless?

"The poll of hypothetical match-ups for 2016 was conducted Aug. 26 to Sept. 9 of 1,115 registered voters. It has a margin of error of +/- 2.9 percentage points. The second match up polled 432 registered Latino voters. It has a margin of error of +/- 6.2 percentage points."

http://www.msnbc.com/msnbc/poll-democrats-claim-resounding-latino-support-over-gop

I might get a 30+ point edge!

Thank you in advance.

Please see post 11 above

?

We are doing our best to get his words in Spanish and into the ears of my gente.

On the other hand, I love Castro and hope she sticks with him. If the GOP selects Bush, then Castro will be 100% required, since we will see Mrs. Bush suddenly morph into full on Carmen Miranda.

He and Hillary fit well together and compliment each other well.

Clinton/Castro 2016 will look great on bumper sticker!

And, no --- I'm NOT saying HC is 'Daffy Duck'.

Clinton 50%, Rubio 44%
Clinton 49%, Bush 45%
Clinton 52%, Cruz 41%

http://www.msnbc.com/msnbc/poll-democrats-claim-resounding-latino-support-over-gop