The 2020 Election Steal: How You’ll Prevent It From Ever Happening Again (Ep.22)
Boeing engineering project manager & process integrity specialist Jim Miller joins us to explain how the 2020 election was stolen and what you can do to prevent it from ever happening again in a few simple steps.
When you login first time using a Social Login button, we collect your account public profile information shared by Social Login provider, based on your privacy settings. We also get your email address to automatically create an account for you in our website. Once your account is created, you'll be logged-in to this account.
DisagreeAgree
Connect with
I allow to create an account
When you login first time using a Social Login button, we collect your account public profile information shared by Social Login provider, based on your privacy settings. We also get your email address to automatically create an account for you in our website. Once your account is created, you'll be logged-in to this account.
DisagreeAgree
3 Comments
Inline Feedbacks
View all comments
Nonya
3 years ago
A lot of these sources appear to be based on the very flawed Electoral Studies paper, which used survey data. This article https://www.cato.org/blog/noncitizens-dont-illegally-vote-detectable-numbers is a good breakdown of why that study is flawed (not really surprising to those who understand maths and statistics!).
Thank you for your input. I tend to believe that referencing a peer reviewed scientific study published in a scientific journal holds more weight than a political opinion piece by Cato, but as always it’s up to the viewer to decide.
For all your readers, the issue with this study is that they didn’t account for measurement error. For example, let’s take vaccination status. About 80% of Americans have gotten at least one shot, and ~10% of the country is hardline anti-vaxx. In this randomized poll (let me note that the CCES survey is not random and is oversampled for citizens), they ask 10,000 Americans if they have been vaxxed, and if they are an anti-vaxxer. The results come back and it shows something weird…predictably, ~1,000 Americans said they were anti-vax, but ~50 of those group said that they got the vaccine! What happened? Well, if you have even a 1% polling error, you’ll get approximately 80 people who “checked the wrong box”…they aren’t anti-vaxxers, and said that they are. So when some of those 80 people check the box saying that they were vaccinated, it makes sense given that they’re actually not anti-vaxxers.
This, of course, is a highly simplified story (what of anti-vaxxers who were forced to get vaxxed, people lying, people who don’t understand the terminology, etc.) but it does actually reflect the problems of analyzing low-frequency events in large sample surveys (including ones on vaccination!).
Any “expert” who brings up this paper with such an obvious statistical error isn’t being good-faith.
A lot of these sources appear to be based on the very flawed Electoral Studies paper, which used survey data. This article https://www.cato.org/blog/noncitizens-dont-illegally-vote-detectable-numbers is a good breakdown of why that study is flawed (not really surprising to those who understand maths and statistics!).
Thank you for your input. I tend to believe that referencing a peer reviewed scientific study published in a scientific journal holds more weight than a political opinion piece by Cato, but as always it’s up to the viewer to decide.
1) Peer review isn’t a synonm for 100% verified, especially at a social science outlook like electoral studies (though perhaps that’s unfair)
2) There are plenty of peer-reviewed articles showing the problem with this study and/or outright refuting its claims: https://www.sciencedirect.com/science/article/pii/S0261379415001420
https://www.sciencedirect.com/science/article/abs/pii/S026137941730166X
https://onlinelibrary.wiley.com/doi/abs/10.1111/psq.12721
3) Your lie detector really should be going off since the side claiming illegal immigrants vote haven’t offered very many cases of this happening.
For all your readers, the issue with this study is that they didn’t account for measurement error. For example, let’s take vaccination status. About 80% of Americans have gotten at least one shot, and ~10% of the country is hardline anti-vaxx. In this randomized poll (let me note that the CCES survey is not random and is oversampled for citizens), they ask 10,000 Americans if they have been vaxxed, and if they are an anti-vaxxer. The results come back and it shows something weird…predictably, ~1,000 Americans said they were anti-vax, but ~50 of those group said that they got the vaccine! What happened? Well, if you have even a 1% polling error, you’ll get approximately 80 people who “checked the wrong box”…they aren’t anti-vaxxers, and said that they are. So when some of those 80 people check the box saying that they were vaccinated, it makes sense given that they’re actually not anti-vaxxers.
This, of course, is a highly simplified story (what of anti-vaxxers who were forced to get vaxxed, people lying, people who don’t understand the terminology, etc.) but it does actually reflect the problems of analyzing low-frequency events in large sample surveys (including ones on vaccination!).
Any “expert” who brings up this paper with such an obvious statistical error isn’t being good-faith.