Donald Trump’s Strongest Supporters: A Certain Kind of Registered Democrat...43% of them...

originally posted by: Phage
a reply to: UKTruth
Perfect. That makes sense.

So to say that 43% of those who favored Trump were registered Democrats the wrong way to look at it.

Yes, totally the wrong way to look at it. Even though we don't know the total number of democrats that were 'leaning' polled, it HAS to be quite low because of the percentages within each group. There is no scenario you could come up with that would have a high number polled at 43% support from Trum and still be able to get to the overall 33% average.

Given my scenario above, which is likely to be quite close to the mix (for a number of reasons I wont go into) the actual percent of his overall support from democratic leaners is going to be around 18%.

Obviously the support he has from ALL registered democrats can NOT be ascertained from this data.

originally posted by: Vector99

originally posted by: Phage
a reply to: Vector99
www.abovetopsecret.com...

Mkayyy so do you validate this poll without raw data?

No.

originally posted by: Phage

originally posted by: Vector99

originally posted by: Phage
a reply to: Vector99
www.abovetopsecret.com...

Mkayyy so do you validate this poll without raw data?

No.

For once we are in agreement.

a reply to: Vector99

For once we are in agreement.

No, not really. You have said that the poll is invalid. I have not.

I have said that whether or not the poll is valid, the OP's interpretation of the results is invalid.

a reply to: Phage

You can either validate or invalidate a poll. No in between.

Do you validate this poll or not? Not speaking opinion, speaking statistical data. What is your take?

a reply to: Vector99

You can either validate or invalidate a poll. No in between.

Says who? Why do I have to do either one?

Do you validate this poll or not? Not speaking opinion, speaking statistical data. What is your take?

That is a meaningless question. How can I do either without access to the data?

Why do you constantly make me repeat myself to you?
I have said that whether or not the poll is valid, the OP's interpretation of the results is invalid.

a reply to: Phage

There isn't raw data, That is how you decide.

originally posted by: Phage
the OP's interpretation of the results is invalid.

I never thought it was.

a reply to: Vector99

There isn't raw data, That is how you decide.

Ok. You go ahead and base your opinion on no information.
That's not my mode of operation.

originally posted by: matafuchs
a reply to: Phage

1,500 were registered Democrat saying they were leaning Repblican and 43% said Trump, so 645 said Trump

Forty Three Percent...

Its not really that difficult to understand Phage's point.

e.g.
A : 10000 of the people called say they are registred democrats.
B : 1500 of them say they are leaning towards the Republican party
C: 43% of them, or 645 say Trump

The scenario above would mean that 6.45% of registered Dems overall were leaning to Trump.

Unfortunately we don't know the value of A at all and can't estimate it
For B we can take an educated guess
We know C.

The only scenario where 43% of registered Dems favour trump is if 100% of all Dems called said they were leaning towards the republicans! If that were the case, thsi survey would have been much bigger news on every media outlet all over the world.


EDIT:
ACTUALLY there is a way of estimating A. Phage has already found that about 41,000 people were telephone interviewed and we know from other surveys that on average. About 67% of US citizens are registered and c45% of those are registered Dem. so about 30% of sample is likely to have been registered DEM, or 12,000 people (not far off my scenario of 10k). Given B above, it would suggest 5% of rgeistered DEMS overall would be leaning to Trump. The value of B is a scenario of course, but if you play with the numbers you can get to a pretty solid conclusion that the range would be somewhere from 2% - 8%.

a reply to: Phage

But you did. You based your opinion on the article.

a reply to: Vector99

But you did. You based your opinion on the article.

Once more. Then I am finished.

I...have...no...opinion...on...the...validity...of...the...poll.

originally posted by: Phage
a reply to: Vector99
Once more. Then I am finished.

I...have...no...opinion...on...the...validity...of...the...poll.

Sorry to irritate you great sir, but the ONLY thing I have tried to argue is the results of this entire study are invalid. They don't source any data. You chose to support some parts while at the same time denying some. You cherry-picked the article and data, not me. I told you from the start it is complete BS. Yes I will invalidate the complete study simply because they do not source their data. That is enough for me to discredit something.

a reply to: Vector99

They don't source any data. You chose to support some parts while at the same time denying some.

I am responding now because you have stated a falsehood about my position and I will not let that stand.

I have never supported the validity of any part of the information provided nor have I denied any of it. I questioned as to how some of the analysis may have been arrived at and my question was answered (not by you) satisfactorily.

a reply to: Phage

If you are not supporting it or denying it...what is the point of your argument?

originally posted by: matafuchs
a reply to: Phage

If you are not supporting it or denying it...what is the point of your argument?

I think he was trying to advise you that in no way shape or form does this survey say that 43% of registered dems are leaning towards Trump. At least that is what I read from his posts.

a reply to: Phage

originally posted by: Phage
a reply to: matafuchs

It does not include people who are undecided either.

How do you know this?

I just noticed something though. For the 33% who favor Trump:
Unregistered: 40%
Republican: 29%
Independant: 36%
Democrat: 43%
See anything odd?

Ok Phage, you win. You didn't omit asterisk data here, nor did you look into it. I concede.

a reply to: UKTruth

it would suggest 5% of rgeistered DEMS overall would be leaning to Trump. The value of B is a scenario, but if you play with the numbers you can get to a pretty solid conclusion that the range would be somewhere from 2% - 8%

A bit of a clarification is required. The poll only asked which of the candidates would be favored by the subjects in a Republican primary. This does not indicate that they would vote for Trump rather than someone in their own party. This is also an unknown since no Democrat candidates were presented as an option.

originally posted by: UKTruth

originally posted by: matafuchs
a reply to: Phage

If you are not supporting it or denying it...what is the point of your argument?

I think he was trying to advise you that in no way shape or form does this survey say that 43% of registered dems are leaning towards Trump. At least that is what I read from his posts.

Yes. I have been arguing one thing and one thing only. That the OP's interpretation of the data as presented is invalid.

originally posted by: Phage
a reply to: UKTruth

it would suggest 5% of rgeistered DEMS overall would be leaning to Trump. The value of B is a scenario, but if you play with the numbers you can get to a pretty solid conclusion that the range would be somewhere from 2% - 8%

A bit of a clarification is required. The poll only asked which of the candidates would be favored by the subjects in a Republican primary. This does not indicate that they would vote for Trump rather than someone in their own party. This is also an unknown since no Democrat candidates were presented as an option.

Yes agreed - somewhere between 2-8% of of all regsitered DEMs would favour a vote for TRUMP amongst all the REPUBLICANS ,IF AND ONLY IF they decided to follow through on their leaning away from the DEMS to the REPUBLICANS.

Thats about the best you can do with the data, and even then you have to sprinkle in some creative data licence.