I remember there was a very short time when he seemed serious and was talking rationally before he flipped back to attacking states, talking about opening by Easter and all that this is probably a reflection of the change in tone back and forth.
You don't need to be a statistician to know how sample sizes work lol.
I know you're getting piled on, but let me try to actually explain this point.
If we were able to pick truly at random, i.e. if every registered voter in the US had an equal chance to be picked for this survey, math can tell in very accurate terms how likely it is for our results to be different than the actual opinions of total population.
It's not a single number, it's probability curve - it is something that can tell us things like "there is 95% chance that our results are within 5%" (this is often reported as margin of error but that figure alone is always incomplete). There can always be outliers, these are stats, you can be 50% off, you can flip a coin and it will land on its head 100 times in a row. It's not likely, but it's possible. For 200 million people, in a sample that size, if it was truly random you would have about 99% chance of being within 5%. I know it may sound counterintuitive, but that's math for you.
Now you may ask, why not just increase the sample size and get even more accurate?
That is because the main source of error in polling actually comes from sampling bias, in practical terms what it means that some people are more likely to answer polls than others. Increasing the sample size doesn't solve this because the bias just remain. This is a difficult problem that can never be solved perfectly, but increasing the sampling size doesn't actually make it easier to tackle, it usually make it harder.
Hope this helps.
Probably that and seeing him extorting several states for medical supplies while people are dying.I would imagine that the two or three days of insane "sacrifice yourself for the economy" messaging contributed to this. I actually believed at first that the GOP base was brainwashed enough to go for it too LOL
I would imagine that the two or three days of insane "sacrifice yourself for the economy" messaging contributed to this. I actually believed at first that the GOP base was brainwashed enough to go for it too LOL
Trump to reporter asking about ventilators: Don't be a cutie pie — CNN Politics
During a White House coronavirus task force briefing, President Donald Trump responded to a reporter's question asking for a guarantee on ventilator availability saying "don't be a cutie pie."apple.news
But that also shows how worthless these polls actually are. When your sample size is 0.0002% of the US population, how is this even remotely close to accurate?
https://www.coursera.org/learn/basic-statistics
To be fair that particular reporter was trying to bait him. What a stupid question to ask. He was obviously trying to get Trump to say something stupid to spin another news story.At yesterday's conference, Trump called certain reporters "wise guy" and "cutie pie" when he was being grilled.
The only thing Trump has working for him is that the rural areas with a lot of his voters will be those with a relatively smaller impact. You can see the start of the narrative already by some deranged people that "god is hitting blue states harder than red ones".
It'll get worse when those deaths start happening close to home and people realize how serious this is.
The only thing Trump has working for him is that the rural areas with a lot of his voters will be those with a relatively smaller impact. You can see the start of the narrative already by some deranged people that "god is hitting blue states harder than red ones". I don't even want to imagine the effect the crisis will still have on the election itself when people may still be scared to go out and vote. Combine that with the stupid idea of an electoral college and I see a lot of things actually working in Trump's favor.
What will be the narrative in a week when Louisiana, Texas, Florida, and Mississippi all equal or surpass NYC's numbers?
This seems to be a common sentiment and it's just fucking revolting.Yeah...everyone here at work think the restrictions should be lifted in a week since "it's been two weeks and we know who is sick and who aren't now". People just want it to be over with already.
Florida has many counties on lock down and the numbers haven't bloomed. The vast majority of cases are a Broward and Miami-Dade.What will be the narrative in a week when Louisiana, Texas, Florida, and Mississippi all equal or surpass NYC's numbers?
But that also shows how worthless these polls actually are. When your sample size is 0.0002% of the US population, how is this even remotely close to accurate?
Florida has many counties on lock down and the numbers haven't bloomed.
Florida has many counties on lock down and the numbers haven't bloomed.
I know full well how many cases. Changes nothing I said.
None of this "haha lib states" grandstanding is going to matter by mid-April, everyone will be begging for supplies and wondering aloud why we're not ahead of this thing.
Again. Check the county map. And the counties have taken care of stay in lock ins themselves.It's that state's governor that hasn't been doing a great job at doing that. He hasn't closed down the boarders to his state, & hasn't gotten Trump on the phone to halt all incoming flights to Florida.
It may be small right now, but check back in a week. It's only going to get worse.
That's a really cool way of explaining it. I'm gonna remember that for sure.it can be very accurate assuming it's a representative sample of the population, that's just how statistics works
it's like how you don't have to drink an entire gallon of a beverage to know how it tastes
assuming it's the same the whole way through, all it takes is a sip to know if you like it or not
these polls attempt to "take a sip" by surveying a group of people that looks like the larger population statistically
His handlers really need to get him off these daily briefings.
Every survey/statistics thread, without fail.
Did most high schools not do statistics at some point? I think we did it in 9th grade.
Thanks for taking the time to explain this and not just be a total asshole.I know you're getting piled on, but let me try to actually explain this point.
If we were able to pick truly at random, i.e. if every registered voter in the US had an equal chance to be picked for this survey, math can tell in very accurate terms how likely it is for our results to be different than the actual opinions of total population.
It's not a single number, it's probability curve - it is something that can tell us things like "there is 95% chance that our results are within 5%" (this is often reported as margin of error but that figure alone is always incomplete). There can always be outliers, these are stats, you can be 50% off, you can flip a coin and it will land on its head 100 times in a row. It's not likely, but it's possible. For 200 million people, in a sample that size, if it was truly random you would have about 99% chance of being within 5%. I know it may sound counterintuitive, but that's math for you.
Now you may ask, why not just increase the sample size and get even more accurate?
That is because the main source of error in polling actually comes from sampling bias, in practical terms what it means that some people are more likely to answer polls than others. Increasing the sample size doesn't solve this because the bias just remain. This is a difficult problem that can never be solved perfectly, but increasing the sampling size doesn't actually make it easier to tackle, it usually make it harder.
Hope this helps.
Don't mention it.Thanks for taking the time to explain this and not just be a total asshole.
The only thing Trump has working for him is that the rural areas with a lot of his voters will be those with a relatively smaller impact. You can see the start of the narrative already by some deranged people that "god is hitting blue states harder than red ones". I don't even want to imagine the effect the crisis will still have on the election itself when people may still be scared to go out and vote. Combine that with the stupid idea of an electoral college and I see a lot of things actually working in Trump's favor.
I like Qualtrics a lot as a service, but this is not helpful.Sample Size Calculator - Qualtrics
Need to quickly calculate sample size? Use our calculator to find your ideal sample size now! ✓ Learn more with this complete guide.www.qualtrics.com
As I expected, recent polling were outliers/over sampling enthusiasm for the Trump administration's *slightly better* handling of the situation. Reality has struck back in.
I'm going to assume those that approve are voting for him regardless of what he does.
True.Which is very comforting. The more people disapprove, the better I feel. I have no idea if that's a healthy mindset, but hey.
That's why these "statistic" threads always make me laugh.