Monday, October 24, 2016

Leftism and Data Manipulation

New Podesta Email Exposes Playbook For Rigging Polls Through "Oversamples"

Earlier this morning we wrote about the obvious sampling bias in the latest ABC / Washington Post poll that showed a 12-point national advantage for Hillary. Like many of the recent polls from Reuters, ABC and The Washington Post, this latest poll included a 9-point sampling bias toward registered democrats.
"METHODOLOGY – This ABC News poll was conducted by landline and cellular telephone Oct. 20-22, 2016, in English and Spanish, among a random national sample of 874 likely voters. Results have a margin of sampling error of 3.5 points, including the design effect. Partisan divisions are 36-27-31 percent, Democrats - Republicans - Independents."
Of course, while democrats may enjoy a slight registration advantage of a couple of points, it is nowhere near the 9 points reflected in this latest poll.

Meanwhile, we also pointed out that with huge variances in preference across demographics one can easily "rig" a poll by over indexing to one group vs. another. As a quick example, the ABC / WaPo poll found that Hillary enjoys a 79-point advantage over Trump with black voters. Therefore, even a small "oversample" of black voters of 5% could swing the overall poll by 3 full points. Moreover, the pollsters don't provide data on the demographic mix of their polls which makes it impossible to "fact check" the bias...convenient.

Now, for all of you out there who still aren't convinced that the polls are "adjusted", we present to you the following Podesta email, leaked earlier today, that conveniently spells out, in detail, exactly how to "manufacture" the desired data. The email starts out with a request for recommendations on "oversamples for polling" in order to "maximize what we get out of our media polling."
I also want to get your Atlas folks to recommend oversamples for our polling before we start in February. By market, regions, etc. I want to get this all compiled into one set of recommendations so we can maximize what we get out of our media polling.
The email even includes a handy, 37-page guide with the following poll-rigging recommendations. In Arizona, over sampling of Hispanics and Native Americans is highly recommended:
Research, microtargeting & polling projects
- Over-sample Hispanics
- Use Spanish language interviewing. (Monolingual Spanish-speaking voters are among the lowest turnout Democratic targets)
- Over-sample the Native American population
For Florida, the report recommends "consistently monitoring" samples to makes sure they're "not too old" and "has enough African American and Hispanic voters." Meanwhile, "independent" voters in Tampa and Orlando are apparently more dem friendly so the report suggests filling up independent quotas in those cities first.
- Consistently monitor the sample to ensure it is not too old, and that it has enough African American and Hispanic voters to reflect the state.
- On Independents: Tampa and Orlando are better persuasion targets than north or south Florida (check your polls before concluding this). If there are budget questions or oversamples, make sure that Tampa and Orlando are included first.

Meanwhile, it's suggested that national polls over sample "key districts / regions" and "ethnic" groups "as needed."

- General election benchmark, 800 sample, with potential over samples in key districts/regions
- Benchmark polling in targeted races, with ethnic over samples as needed
- Targeting tracking polls in key races, with ethnic over samples as needed
I don't get it. Why even make the phone calls and actually take data? Why not just make it up from scratch and avoid all that work? Unless, of course, you want to feed the REAL data to the Clinton campaign for Hillary's use. Otherwise, creating a phony 12 point lead could work against Hillary, as voters assume that it's a done deal for her election, and so they stay home. I presume that the Left thinks that Republicans would stay home in higher numbers. But it's a two edged sword. I read somewhere a few days ago that the polls always seem to converge on the real real voter expectations when it gets very close to election time. That could be real, or it could be cover-up for earlier data manipulation.

22 comments:

Robert Coble said...

The calls are made to provide plausible deniability in the extremely unlikely event that anyone ever actually looks at the raw data.

"There are lies, damned lies, and then statistics (polls), in order of shamelessness."

Are you expecting "truth telling" from "scientific" shills who are PAID to lie for the highest bidder?!?

Steph said...

Old whites are over represented in polls. They're trying to compensate for that. It's clear from sentences such as "Consistently monitor the sample to ensure it is not too old, and that it has enough African American and Hispanic voters to reflect the state."

To reflect the state. Not to make it look inflated.

Stan said...

If that's the case, then the polls are not random in their construction and implementation, and cannot in any manner represent the actual voting population. It's just like AGW in that it requires adjustment in order to actually reflect the narrative which it supposedly proves.

This is not manipulation being mathematically required in order to be accurate; there is no mathematical reasoning involved, nor demographic reasoning either.

It merely is the Democrat standard approach to everything involving information: corrupt the information in a manner which either does not reflect negatively on Democrats, OR at least makes non-Democrats look bad, OR is to be suppressed altogther. That is SOP for the MSM, the Leftist government, and the Democrat Party.

Corruption killed Bernie's candidacy. Corrupting Trump's statements regarding minorities is common right up to Hillary Herselfness. Hillary and the Democrats lie, even when they don't move their lips. And it works for the leftist True Believers, who either blindly buy it, or are corrupt themselves.

The truth will be known in two weeks.

Stan said...

I just remembered: they need to count the number of out-of-state voters who are driven in, handed fraudulent credentials and paid to vote Democrat.

That sort of behavior would, in fact, require an adjustment to the data, assuming that the data is properly acquired. That doesn't preclude "old whites" from the population of illegal voters, though. I imagine Bernie would do it, if he need the $20 though.

Stephh said...

This ..is.. manipulation being mathematically required in order to be accurate
..and..
Geez dude, are you auditioning for Limbaugh's job?

Stan said...

So you think that Podesta is a non-aligned, unbiased mathematician? That Podesta is telling the poll designers how to "unbias" their polls? Really?

I don't listen to Limbaugh. I think things through. And believe me, considering Podesta to be an unbiased mathematician who specializes in statistics and honest data in order to prevent bias, is not rational.

Stephh said...

No idea, was just quoting...
You infer so much from so little

Stan said...

You actually don't know who Podesta is, what he has done in the past, and who pays him, do you?

Stephhn said...

As I said
No idea, was just quoting...

Stan said...

It is customary to give the source of the quote you are using. That way it is obvious that it is not your own words. However, I doubt that the crack about Limbaugh was a "quote", and therefore your overall credibility has taken a serious hit.

Stephhn said...

option 1. polling samples are not completely random as to reflect the state's demographic, this shows Trump has only about 15% chance of winning based on expected electoral college votes.

option 2. everything is garbage, everyone is lying, polls are rigged, the election is rigged, lefties are out to get us, the world is about to end if crooked Hillary gets in, but real americans will vote Trump and there are more of us but dems and the evil corrupt msm are trying to hide that just like global warming is a hoax because math is corrupted and that's all related y'know cause we're the rational ones but others aren't

Stan said...

[Your name is morphing, are you evolving?]

Your "options" are caricatures, and as such they are intended to be arguments which are ridiculous in your eyes, and yet are not arguable rationally because they are caricatures, and not real arguments.

But let's discuss them anyway.

Option 1. polls are intended to be evidence of some sort. But the evidence is historically seen to be corrupt. What is more interesting is the motivation for the corruption. Believing the polls to be accurate is not rational given the spread of their conclusions. Adding the margin of error to the spread gives roughly +/- 20%, and that applies to a race that is likely a dead heat. The polls absolutely cannot be believed.

Option 2. Here your caricature goes off the rails. Not everyone is lying. The Democrats have been demonstrated to have been lying, buying violence at Trump rallies, and investing in dirty tricks of all sorts. Trump? Name one.

Assuming that the nation elects Hillary, the corruption will be institutionalized in the Executive and Judicial Branches of US government for the next 4 to 5 decades. It's easy to see, if you choose to actually step back and view all - ALL - the evidence which has been produced.

And of course, AGW. how many times must the original data be "adjusted" in order to support the desired hypothesis? Why is AGW a "vital science" when it cannot be falsified, like all other science pursuits (except evolution)? Why would scientists who do that in non-ideological pursuits NOT be relieved of their credentials? Answer those issues, OK?

So far, you have merely snarked at issues, and not produced any meaningful information. Try a little harder.

Stan said...

It's up to you to show that he is lying.

Stephan said...

Are you kidding!?!? What about his Miss Universe thing where he bragged about walking in the changing room, and then said he never did because it's grossly inappropriate?

Just check that out for examples, and there's some from both candidates obviously:
https://www.factcheck.org/person/donald-trump/

Steph said...

Stan already showed methodology doesn't matter, facts don't matter, everything is going to hell because he thinks so, nothing can prove him wrong. If facts disagree with his opinion, there must be a liberal bias behind these facts, some cover-up by the MSM, or something like that. What matters is who bring up the point, not what the point is. See his response above to my comment on polling.

Stephan said...

That doesn't really matter. The main point here is that Trump is lying, a lot, yet Stan ignores that it seems!!!

Steph said...

No you're wrong. What matters is -who- is talking. Trump lies don't matter. Excuses will come up anyway. Trump is on the right side of the issues they care about, so whatever he says on the side is just for show. He gets attention by being provocative.

Stan said...

For information on fact checker.org, go here:
http://www.matchdoctor.com/blog_141905/Factcheck_org_--_A_Fraudulent_Fact_Check_Site_Funded_By_Biased_Political_Group.html

Stephan said...

Wow this blog sounds familiar... like word for word familiar. Like some sort of propaganda written specifically to distribute across multiple blog sites familiar. Like it's written a group who's job it is to write things to dispute reputable news organizations or something so idiots have something to reference when someone points to a factcheck.org article.

Oh! that's exactly what it is! Because this blog post exists word for word on all these sites!

http://tinyurl.com/coy2y5m

Stan said...

"At least she knows what she's doing and won't make a joke out of our country."

The USA is already a joke. Clinton commits treason and walks. She's married to a serial rapist and child rapist, who gets paid $500k per speech as investments for pay to play, and walks. They plundered Haiti and walked. She virtually created Boko Haram and ISIS, and half the country thinks she's wonderful. Half the country thinks that is just A-OK, because they think that they can get in on the gov't dole too, because she says she will "invest" in them. Just like the Dems invested in blacks just enough to keep them "in their place" for the past 60 years. And half the country is fine with that.

The Democrats send paid thugs to rough up Trump followers, and then blame Trump for the violence. Half the country is fine with that.

Yes, the USA is now a sorry sick joke.

Anyone who prefers treason and traitors and racists and rapists as his POTUS is just as morally corrupt as they are.

Steph said...

Ya that sounds almost right, except that you're completely off the chart and have nothing to back up such grand claims, you show no respect for the institutions that made our country so great, just because they dont support your opinions, and you completely ignore Trump's scandals and inexperience, which makes him a worse candidate even when recognizing all of Clinton's actual flaws. Typical alt-right BS, the SJWs of the Right.

Stan said...

You are merely a troll. It's obvious by all the internal contradictions which you make in just the first sentence, which is actually most of the paragraph.

You'll make no more comments here.