Trump hires pollster who says CNN skewed survey for low approval rating

WASHINGTON — President Trump on Monday hired a pollster to scrutinize a CNN poll that pegged him at an approval rating of just 38 percent in the wake of historic civil unrest, his worst since January 2019.

The analysis from long-time GOP pollster John McLaughlin, which Trump shared on Twitter, accused CNN of deliberately under-polling Republicans and those likely to vote in what he claimed was an attempt to suppress the president’s supporters.

“They continue to poll adults or registered voters that skew away from likely voters,” McLaughlin wrote in the letter addressed to Trump on Monday.

A CNN poll, perhaps it is not trustworthy.

More suggesting the polling was flawed.

Is anyone surprised? Its all about poll numbers and ratings to this guy. Even when FOX tells him he’s trailing, he bad mouth’s them.

If the pollster doesn’t come up with him leading by at least 10%, it makes you wonder if he will sue that poor person too…because in Donald Land, a friend one day is an enemy tomorrow!

In another thread about Trump’s hostility to the free press, his defenders were demanding some concrete action Trump took against the free press as evidence. Here. Here in this legal action and mudslinging campaign is a concrete action Trump has taken against the free press.

The poll was conducted by the SSRS. If you want to know about the polling methods check them out.

The things that could be appended to “perhaps” are only bound by the imagination. Not that such musings are necessarily informative or necessarily concise with fa ts or information.

Strong economic recovery?

There are almost 40 million unemployed.

2 Likes

Back to the discussion of the article:

Per McLaughlin argument - What do you all think of a poll that was comprised and represented by approximately 25% republican, 32% democrat and 44% independent and other party representation? I find the position valid, I usually look at polls that dont ask questions with built in bias and that represent the electorate in ratios to the voter demographic [D, I, R] …and I think likely voters is far more predictive than registered voters.

The mere fact that they under represented republicans by a large amount lends credence to the claim of its being used as a suppression poll … of course to that end - it is 1] being used to gin up support for Biden and/or to attempt to show that Trump is going to loose - ergo suppress and dishearten the republican vote

I really like when they publish the questions - because you can tell whether it was a push poll … you know - the kind that has questions like"

"How concerned are you that former Vice President Biden inappropriately touches young women?’ Extremely concerned 10 … not concerned at all 1

“How concerned are you that President Trump will not leave the White House if he losses the election?” Extremely concerned 10 … not concerned at all 1

I stopped caring about polls and what they report. Most appear to be wrong with their predictions.

I can remember too 4 years ago when Hillary Clinton’s campaign managers e-mail account was leaked he was saying to get skewed polls out. It makes sense for someone in the business of politics to do this. Polls can influence people.

I recall seeing this article yesterday too, I’ve seen similar articles over the last few years, about polling and how they can be off due to biases.

Doug Schoen: Why Trump-Biden race is much closer than you think

1 Like

Without hearing more about the statistical methods used the apparent premise that this distribution of political orientations compromises the survey is not agreed upon.

How many republicans do you think there are in the US? This is Gallup.

Edited to add: For a while, the electorate’s response to the parties has been, “A pox on both your houses”.

Many business are still closed. Is that Trump’s fault?

1 Like

There are 40 million unemployed.

That means the economy is doing rather lousy.

I didn’t mention Trump.

1 Like

No one has really addressed what the article discuses about they the author believes the polls are problematic - one of which is that republicans were under represented. You a have not either …

So, are you saying they were not based upon what you posted? Because even what you posted illustrates an under representation.

Pollsters had Hillary winning in 2016, even up to election day … There were only a couple of polls that had Trump with a possible win … and the same D/R disparity existed in many of the 2016 polls.

I am not a pollster - However, Historically polls using:
“Likely voters” perform better than polls where “registered voters” are used
Parity between the parties - where the numbers are not skewed one way or the other

Are the best predictors of real outcomes

This far out from the election … probably not as important anyway

Would be nice for someone to address/refute the points made in the article

How to Lie With Statistics by Darrell Huff

1 Like

Didn’t Congress just allocate $500 billion to fix this? Not enough?

1 Like

Nate Silver had Trump’s odds at 29%. Better than the Cubs had when they were down 3-1 against Cleveland about the same time.

The nation is about 1/3, 1/3, 1/3 between GOP, Dem and independent, but that varies across the population, and varies depending on whether you want all adults, registered voters, likely voters, or some other model. Many pollsters will attempt to adjust their sample to adjust for variances from what they believe their target population actually looks like, but its a two edged sword – if you leave in the sample bias you are accused of having a bad sample, if you correct for sampling errors you are accused of deliberately manipulating your results. And, of course, its always possible you are wrong about the make up of the population you are measuring and that your sample is actually accurate. A lot of complex math and experience goes into those decisions, but there is still a lot of uncertainty mixed in to it.

The bottom line is that this is one poll. There is nothing I can see about the poll that would say it is a particularly good or bad poll. It puts Trump a little farther behind them some other polls, but it is broadly in the range of what most reputable pollsters are finding right now.

I have victoria33 on ignore, so I can’t see the original article.

You said “large amount”. What I posted shows it isn’t a “large amount”.

And, more people voted for Hillary. So, the national polls were correct.

I haven’t read the methodology, but I assume that the pollster knew the sample bias and accounted for it.

1 Like

A new Isos/Reuters poll put the spread among registered voters at Biden +13

An AbacusData poll put Biden at +14 among registered voters

https://abacusdata.ca/biden_trump_2020_abacus-data/

perhaps the simple truth is the polls are an indication of,…

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.

DISCLAIMER: The views and opinions expressed in these forums do not necessarily reflect those of Catholic Answers. For official apologetics resources please visit www.catholic.com.