The tide is turning…

Is this another poll of 500 people at a stuckies on I10 in Texas?

https://mclaughlinonline.com/pols/wp-content/uploads/2021/09/MA-National-Survey-Release-September-2021.pdf

Nope, just a nationwide sampling of 1000.....not accurate still....

That’s how polls work.

It’s just a snapshot in time of the opinions of those particular 1000 responders.

This is a particularly interesting result though, considering how the question was worded.

“I used to dislike Trump’s confrontational style, but now that I see how weak Biden was in dealing with the Taliban in Afghanistan, I’m thinking maybe Trump wasn’t so bad.”

28% of the Democrats polled agreed with that statement.

Not with a sample size of 1000 to model 328 million. Statistics dont work like that......

It is very important to use a correct sample size. When your sample is too big, this will lead to unnecessary waste of money and time. On the other hand, when it’s too small, your results will not be statistically significant and you will not come to reliable conclusions.

Meh… I really don’t get too worked up about polls anyway, and I make no claim that it accurately represents the nation.

But, that particular question and answer really caught my eye.

It’s just a snapshot in time of the opinions of those particular 1000 responders, but I found how those particular 1000 people chose to respond to that question to be quite interesting.

Given the results on that question, I can certainly understand why you’d want to talk about sampling sizes and statistical significance instead.

Statistics hard....

“I used to dislike Trump’s confrontational style, but now that I see how weak Biden was in dealing with the Taliban in Afghanistan, I’m thinking maybe Trump wasn’t so bad.”

Tattoo it across your back!

I don’t do tattoos.

But, if I did…

“I used to dislike Trump’s confrontational style, but now that I see how weak Biden was in dealing with the Taliban in Afghanistan, I’m thinking maybe Trump wasn’t so bad.”

would be a little too wordy.

Actually, they do. This article is old, but I doubt statistics have fundamentally changed since then. If I'm wrong about that, I'd be happy to admit it, once proven.

https://www.scientificamerican.com/article/howcan-a-poll-of-only-100/

I think the one thing the polling organizations haven't figured out how to compensate for, is people just flat out lying. We saw in the last two presidential elections that the polls were way off, because people liked Trump, but didn't want to admit it to a stranger on the phone due to the social stigma the media placed on people who supported Trump. That was more the case in 2016 than 2020, but it was clearly a real thing.