Speaking of bias, the latest from The Claw News, a subscription newsletter (https://theclawnews.com) had this observation about a recent 538 polling story.
(ROC stands for Rule Of Claw - and the newsletter is enlivened with cat memes. I've added links to this, since they didn't cut and paste from the original newsletter.)
Speaking of bias, the latest from The Claw News, a subscription newsletter (https://theclawnews.com) had this observation about a recent 538 polling story.
(ROC stands for Rule Of Claw - and the newsletter is enlivened with cat memes. I've added links to this, since they didn't cut and paste from the original newsletter.)
"The Claw News October 27th, 2021
A Shockingly Biased Republican Push Poll Is Incorporated Into 538 Averages. See The Proof HERE.
538 is the ultimate aggregator of political polls both valid and invalid, and somehow or another, they have a special sauce that assigns weights and creates adjusted leaders based on how biased they think one poll is versus another.
They seem to think themselves not just a publisher of statistics, but the ultimate arbiter of them. The problem is polls have long since moved beyond data collection and into the realm of propaganda. For my part, I do not believe in weighing already weighted statistical samples. Data is data. But when is data not data? When the collection of data manipulates the interviewee to the point where the entire collection is rendered invalid, or what we call a push poll.
I am not saying 538 is attempting to put their thumb on the scale. I am saying that in my opinion they should be more restrictive. They rate the poll in question as C/D which is not good, obviously, but that does not tell the whole story. If you allow skewed data into your averages, your averages themselves are skewed, no matter how much you try to adjust data collected with an agenda.
In order to fulfill Trump’s demand to discredit CNN’s poll, he needed to bring in an outsider capable of generating even more outrageous claims. Hence the appearance of “highly respected pollster” John McLaughlin, who is, in fact, a laughingstock within the profession.
538 can weight all it wants, but if the method of collecting data is biased the entire process is invalid..how could the results ever be valid?
Let me ask you a question-
“Many people think that Ford makes ugly, unreliable overpriced vehicles.”
Do you agree?
This question can introduce a response bias, or less likely a reverse courtesy bias.
Now look at this question-
As you are likely aware Ford is an American auto manufacturer. Is your opinion of Ford
1. Positive
2. Negative
3. Neutral/No opinion.
Simple question. Ford exists. What do you think of Ford?
Not, “Ford sucks.” Do you agree?
So let’s look at this so-called poll’s questions. Because these questions are so shockingly biased, I can not say for sure their question order, which could also greatly change the response validity.
Here is a sample of awful questions-
From what you have seen so far, are you worried about the future of America with Joe Biden as President?
All things being equal, do you prefer:1. Free Market Capitalism or. 2. Big Government Socialism?
I pause here to notate that this pollster got a response of 61 percent in favor of free market capitalism. But they were not satisfied, so they literally prove my whole point with the second question related to the topic.
All things being equal, do you prefer:1. American Capitalism or 2. Big Government Socialism?
This question received 76 percent in support.
But still not satisfied, they ask the question a third time, like this:
All things being equal, do you prefer…? 1. American Opportunity Society OR, 2. Big Government Socialism?
This got them to 78 percent support. So they are message testing propaganda to back capitalism. I get that. Let’s try another hypothetical question proposed by me to make a comparison:
We asked 1,000 consumers which do they prefer?
1. A souped up Dodge Charger
2. A 4 Cylinder Toyota Camry
We found 55 percent preferred the Charger.
When we asked it like this:
1. A souped up Dodge Charger with leather seats
2. A 4 cylinder Toyota Camry
We found 64 percent support for the Charger.
Finally when we asked it like this:
1. A souped up Dodge Charger with leather seats and a free Bose sound system
2. A 4 cylinder Toyota Camry
We found 75 percent of respondents preferred the Charger.
Conclusion?
75 percent of Americans prefer a Dodge Charger to a Toyota Camry.
But the questions continue to get worse in the McLaughlan poll.
Do you approve or disapprove of President Biden’s proposed new federal mileage tax which would tax a driver 8 cents for every mile they drive and cost the average driver over $1200 a year?
This question is not even accurate. President Biden has commissioned a study on various proposals, this being one of them though no amount is outlined. None of them have been formally proposed. (https://apnews.com/article/fact-checking-907285011746)
CLAIM: President Joe Biden has called for a “driving tax” that is estimated to be 8 cents per mile.
AP’S ASSESSMENT: False. The administration has not proposed that tax. A provision in Biden’s $1 trillion infrastructure bill before Congress would establish a national study to assess how such a tax could be implemented. It would not actually enact that tax, nor it does not outline a rate of 8 cents per mile.
This is as close as it gets to a pollster hypothetically asking,
“President Biden wants to personally knock on your door and remove money from your wallet. Do you still like President Biden?”
Yet 538 thinks this poll is worth including?
Next:
Which of the following candidates are you more likely to vote for congress?A REPUBLICAN CANDIDATE who supports allowing school choice for every American child. OR, A DEMOCRAT CANDIDATE who supports more funding for government run schools and raising salaries for government employed teachers
Government run schools? Do you mean public schools? Paid for by tax dollars from people like you and me, with a school board elected by people like you and me with accountability to you and me?
President Biden has urged the FBI to investigate some of the parents who have protested vigorously about Critical Race Theory curriculum at school board meetings.He has said that they may be domestic terrorists. Do you agree or disagree that these parents should be investigated?
The Department of Justice will mobilize the FBI in response to a spike in harassment, intimidation and threats of violence against school board members across the country.
In a memorandum released Monday evening, Attorney General Merrick Garland directed the FBI and U.S. attorneys' offices to meet in the next 30 days with federal, state and local law enforcement leaders to outline strategies for addressing the incidents, which have become more frequent and more volatile in recent months.
1.(Some/Other) people say that President Biden’s weakness in Afghanistan is emboldening China and making them think they can act militarily against Taiwan without fear of retaliation by the United States. OR, 2.(Other/Some) people say China knows that President Biden will stand firm for Taiwan.
Equivalent?
Question A1. asked of participants who identified as male voters:
Some people say eating ketchup will make you impotent. Others say it won’t.
Asking about your approval of President Biden at that point is like..
Are you still willing to eat ketchup?
And after all of that, the poll finds 48 Republicans, 47 Democrats in the generic ballot. This shows just how bad of shape the Republicans are in politically.
Look this is a company that message tests. Fine. Do your thing. But don’t present it as data.
James Carville co-runs with Stan Greenberg Democracy Corps, a political think tank and poll that message tests, though not with this kind of bias so their data is much less suspect.
But 538 takes this McLaughlan poll, sees it is a partisan poll which does not even begin to cover it, and assigns it a three point weight to the Democrats upon review.
So after thinking about it, only 72 percent, not 75, prefer a Dodge Charger to a Toyota Camry.
Thanks for clearing that up.
One more question-some people say 538 is not restrictive enough with the data they use for their poll aggregations.
Thank you. I did look at the list of 538 polls and how they rate them, but did NOT realize that they added these spurious pieces of 'science' to their average.
Speaking of bias, the latest from The Claw News, a subscription newsletter (https://theclawnews.com) had this observation about a recent 538 polling story.
(ROC stands for Rule Of Claw - and the newsletter is enlivened with cat memes. I've added links to this, since they didn't cut and paste from the original newsletter.)
"The Claw News October 27th, 2021
A Shockingly Biased Republican Push Poll Is Incorporated Into 538 Averages. See The Proof HERE.
538 is the ultimate aggregator of political polls both valid and invalid, and somehow or another, they have a special sauce that assigns weights and creates adjusted leaders based on how biased they think one poll is versus another.
They seem to think themselves not just a publisher of statistics, but the ultimate arbiter of them. The problem is polls have long since moved beyond data collection and into the realm of propaganda. For my part, I do not believe in weighing already weighted statistical samples. Data is data. But when is data not data? When the collection of data manipulates the interviewee to the point where the entire collection is rendered invalid, or what we call a push poll.
I am not saying 538 is attempting to put their thumb on the scale. I am saying that in my opinion they should be more restrictive. They rate the poll in question as C/D which is not good, obviously, but that does not tell the whole story. If you allow skewed data into your averages, your averages themselves are skewed, no matter how much you try to adjust data collected with an agenda.
The entire poll I am about to reveal is skewed. Now this poll is brought to us by (full data here https://mclaughlinonline.com/pols/wp-content/uploads/2021/10/National-Monthly-October-Release.pdf) McLaughlan and Associates, a notorious GOP pollster famous for one of its founder’s association with Donald Trump. (https://nymag.com/intelligencer/2020/06/trump-cnn-poll-14-biden-memo-john-mclaughlin.html)
In order to fulfill Trump’s demand to discredit CNN’s poll, he needed to bring in an outsider capable of generating even more outrageous claims. Hence the appearance of “highly respected pollster” John McLaughlin, who is, in fact, a laughingstock within the profession.
538 can weight all it wants, but if the method of collecting data is biased the entire process is invalid..how could the results ever be valid?
Let me ask you a question-
“Many people think that Ford makes ugly, unreliable overpriced vehicles.”
Do you agree?
This question can introduce a response bias, or less likely a reverse courtesy bias.
Now look at this question-
As you are likely aware Ford is an American auto manufacturer. Is your opinion of Ford
1. Positive
2. Negative
3. Neutral/No opinion.
Simple question. Ford exists. What do you think of Ford?
Not, “Ford sucks.” Do you agree?
So let’s look at this so-called poll’s questions. Because these questions are so shockingly biased, I can not say for sure their question order, which could also greatly change the response validity.
Here is a sample of awful questions-
From what you have seen so far, are you worried about the future of America with Joe Biden as President?
All things being equal, do you prefer:1. Free Market Capitalism or. 2. Big Government Socialism?
I pause here to notate that this pollster got a response of 61 percent in favor of free market capitalism. But they were not satisfied, so they literally prove my whole point with the second question related to the topic.
All things being equal, do you prefer:1. American Capitalism or 2. Big Government Socialism?
This question received 76 percent in support.
But still not satisfied, they ask the question a third time, like this:
All things being equal, do you prefer…? 1. American Opportunity Society OR, 2. Big Government Socialism?
This got them to 78 percent support. So they are message testing propaganda to back capitalism. I get that. Let’s try another hypothetical question proposed by me to make a comparison:
We asked 1,000 consumers which do they prefer?
1. A souped up Dodge Charger
2. A 4 Cylinder Toyota Camry
We found 55 percent preferred the Charger.
When we asked it like this:
1. A souped up Dodge Charger with leather seats
2. A 4 cylinder Toyota Camry
We found 64 percent support for the Charger.
Finally when we asked it like this:
1. A souped up Dodge Charger with leather seats and a free Bose sound system
2. A 4 cylinder Toyota Camry
We found 75 percent of respondents preferred the Charger.
Conclusion?
75 percent of Americans prefer a Dodge Charger to a Toyota Camry.
But the questions continue to get worse in the McLaughlan poll.
Do you approve or disapprove of President Biden’s proposed new federal mileage tax which would tax a driver 8 cents for every mile they drive and cost the average driver over $1200 a year?
This question is not even accurate. President Biden has commissioned a study on various proposals, this being one of them though no amount is outlined. None of them have been formally proposed. (https://apnews.com/article/fact-checking-907285011746)
CLAIM: President Joe Biden has called for a “driving tax” that is estimated to be 8 cents per mile.
AP’S ASSESSMENT: False. The administration has not proposed that tax. A provision in Biden’s $1 trillion infrastructure bill before Congress would establish a national study to assess how such a tax could be implemented. It would not actually enact that tax, nor it does not outline a rate of 8 cents per mile.
This is as close as it gets to a pollster hypothetically asking,
“President Biden wants to personally knock on your door and remove money from your wallet. Do you still like President Biden?”
Yet 538 thinks this poll is worth including?
Next:
Which of the following candidates are you more likely to vote for congress?A REPUBLICAN CANDIDATE who supports allowing school choice for every American child. OR, A DEMOCRAT CANDIDATE who supports more funding for government run schools and raising salaries for government employed teachers
Government run schools? Do you mean public schools? Paid for by tax dollars from people like you and me, with a school board elected by people like you and me with accountability to you and me?
President Biden has urged the FBI to investigate some of the parents who have protested vigorously about Critical Race Theory curriculum at school board meetings.He has said that they may be domestic terrorists. Do you agree or disagree that these parents should be investigated?
Stalking and threats of violence against school board members is not “protesting.” (https://www.usnews.com/news/education-news/articles/2021-10-04/doj-to-investigate-threats-against-school-board-members)
The Department of Justice will mobilize the FBI in response to a spike in harassment, intimidation and threats of violence against school board members across the country.
In a memorandum released Monday evening, Attorney General Merrick Garland directed the FBI and U.S. attorneys' offices to meet in the next 30 days with federal, state and local law enforcement leaders to outline strategies for addressing the incidents, which have become more frequent and more volatile in recent months.
1.(Some/Other) people say that President Biden’s weakness in Afghanistan is emboldening China and making them think they can act militarily against Taiwan without fear of retaliation by the United States. OR, 2.(Other/Some) people say China knows that President Biden will stand firm for Taiwan.
Equivalent?
Question A1. asked of participants who identified as male voters:
Some people say eating ketchup will make you impotent. Others say it won’t.
Asking about your approval of President Biden at that point is like..
Are you still willing to eat ketchup?
And after all of that, the poll finds 48 Republicans, 47 Democrats in the generic ballot. This shows just how bad of shape the Republicans are in politically.
Here is the link to their usage in the generic ballot averages on 538. (https://projects.fivethirtyeight.com/congress-generic-ballot-polls/)
Look this is a company that message tests. Fine. Do your thing. But don’t present it as data.
James Carville co-runs with Stan Greenberg Democracy Corps, a political think tank and poll that message tests, though not with this kind of bias so their data is much less suspect.
But 538 takes this McLaughlan poll, sees it is a partisan poll which does not even begin to cover it, and assigns it a three point weight to the Democrats upon review.
So after thinking about it, only 72 percent, not 75, prefer a Dodge Charger to a Toyota Camry.
Thanks for clearing that up.
One more question-some people say 538 is not restrictive enough with the data they use for their poll aggregations.
Do you agree?
-ROC
Thank you. I did look at the list of 538 polls and how they rate them, but did NOT realize that they added these spurious pieces of 'science' to their average.