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

a reply to: Vector99
You have still not provided a way that 148% can be reached.

and they draw data by webcrawling here

No. Their data is obtained by phone canvassing. Clearly stated:

Civis Analytics conducted 40,050 live telephone interviews of adults in the United States contacted on telephones from August 10, 2015 to December 27, 2015. Among respondents of these surveys there were 11,441 self-identified Republican or lean Republican adults. These respondents were asked their candidate preference in the GOP primary. Undecided respondents are not considered as part of the analysis, map, or trend lines.

civisanalytics.com...

a reply to: Phage

They source this

This is in-line with public polling, where 26% of people identify as Republican (and 31.7% identify as Independent): elections.huffingtonpost.com... ↩

link

a reply to: Vector99

Your link doesn't work but it seems to indicate that the phone canvassing is in-line with public polling. What do you think that "in-line" means?

Ah, a link that works. But the question stands.

a reply to: Phage

I can tell you right now a significant majority of people over 18 and under 30 have ditched their landline phones (which these polls use).

On top of that, I -- along with many people I know never answer those polls truthfully or at all. They're annoying. They're disruptive. They always call at dinner time and interrupt family time.

I would be very suspect of polls using cold calls on landlines.

Hell, I usually pretend to not speak English or that I'm some kind of alien or something when I answer them.

a reply to: Phage

I asked this about their data in the other thread

originally posted by: Vector99
a reply to: schuyler

they don't source their dialing data. They conducted 40,500 phone interviews. How many phone calls total did they make? They cited of 40,500 calls of interviewing, 11,441 were self-identified republicans or lean republicans. What is the raw data of that? How many of the rest fell in what category? What is the age group? Where did they call? What about the other 29,059 called. What is their specific and/or preferred political stance?

These numbers do not exist anywhere in their studies. Without raw data, and other raw data to compare it to I can't count it valid. It is way too easy to make a biased poll. I hate polls, they completely contradict the scientific method.

a reply to: MystikMushroom

I can tell you right now a significant majority of people over 18 and under 30 have ditched their landline phones (which these polls use).

The demographics for this poll show that 29% of those included fit that category. Nationwide the figure is less than that. Apparently that demographic is over-represented in this poll, if anything.


I need to rephrase. 29% of those who favor Trump.

originally posted by: Phage
a reply to: MystikMushroom

I can tell you right now a significant majority of people over 18 and under 30 have ditched their landline phones (which these polls use).

The demographics for this poll show that 29% of those included fit that category. Nationwide the figure is less than that. Apparently that demographic is over-represented in this poll, if anything.

The polls have no raw data, anywhere.

originally posted by: Phage
a reply to: matafuchs
No, you have it wrong...again.

Mr. Trump’s share of the vote among self-identified Republicans.

Any idea what 148% of 33% adds up to?

The percentages are of discrete groups.
Adding all those percentages together in relation to the discrete group sizes, I would assume comes to 33%.
This makes sense because it would mean the greatest volume polled would be from the Republican registered group (29% share). All the others shares are higher, so the volumes polled must be lower.

Oh, and I agree its blatantly obvious from the presented data that the 43% relates to registered Democrats who said they were leaning towards the republican party - which actually also means 57% were leaning for another Republican candidate or just generally, but not for Trump (I found that interesting too).

I really don't know why you are having to argue that point again and again

There is no data here that tells us what percentage of overall registered Democratics say they are leaning to Trump.

Overall though I don't like the journalists analysis for the reasons I stated earlier. On the racism metrics, he references the implicit association tests (which would be far more useful) but proceeds to use search volume data alone. Having looked at the map of IAT by state, it shows little to no correlation to Trumps polling percentages. That alone leads me to believe the whole piece is dishonest (even if the raw numbers from the polling company might be correct)

a reply to: UKTruth

The percentages are of discrete groups.
Adding all those percentages together in relation to the discrete group sizes, I would assume comes to 33%.

Four discrete groups, yes. The breakdown is, apparently, of the 33% who favor Trump. I could understand a number less than 100% (if some don't fall into any of those groups). I don't understand greater than 100%.

originally posted by: Phage
a reply to: UKTruth

The percentages are of discrete groups.
Adding all those percentages together in relation to the discrete group sizes, I would assume comes to 33%.

Four discrete groups, yes. The breakdown is, apparently, of the 33% who favor Trump. I could understand a number less than 100% (if some don't fall into any of those groups). I don't understand greater than 100%.

You keep sourcing the article, but also acknowledging it isn't accurate data. I'm confused Phage, are you just debating for debates sake?

a reply to: Vector99
Pardon me?
Where have I acknowledged that it isn't accurate data?

originally posted by: Phage
a reply to: Vector99
Pardon me?
Where have I acknowledged that it isn't accurate data?

originally posted by: Phage
a reply to: Vector99
Fair enough. I would prefer that as well.

But since the OP is using what the article says as the basis of his claim, don't you think it's reasonable to discuss that?
How about this; the OP's interpretation of the claims of the article is incorrect?

Have you found a source for raw data of their polls?

a reply to: Phage

Scenario:

Polling numbers total 11,000 of which

1,000 were unregistered and 40% said Trump, so 400 said Trump
7,500 were registered repbilicans and 29% said Trump, so 2,175 said Trump
1,000 were independents and 36% said Trump, so 360 said Trump
1,500 were registered Democrat saying they were leaning Repblican and 43% said Trump, so 645 said Trump

So overall 3,580 said Trump of the 11,000 sampled = 32,5%

Obviously I dont know the volume splits (the article does not provide them), but a sceanrio like the above is what I have assumed.

a reply to: UKTruth
Perfect. That makes sense.

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

a reply to: Vector99

Have you found a source for raw data of their polls?

No.
Now, please show where have I acknowledged that it isn't accurate data.

a reply to: Phage

Did you ever find raw data of the poll? You said you would prefer that yourself as well.

a reply to: Vector99
www.abovetopsecret.com...

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...

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

Mkayyy so do you validate this poll without raw data?

a reply to: matafuchs
Yes. I know. I saw that.
43% of the self-indicated Republicans who are registered democrats favor Trump.