Except I don’t whine. I simply ask for the science behind the decisions. That upsets you.
Big Ten has Embarrassed themselves.
- Thread starter superfan01
- Start date
No. You definitely whine. "Why can't I go to church ..but people can buy food to eat!?!" Then you get the (obvious) answers and whine out the same rhetorical questions.
Irony is ironic. I wasn't even talking to you,. I dont even think you were in the thread ...lmao.
Also, my post was pure truth. GFY.
Um no, never said why can’t I go to church but people can buy food. But since you brought it up, explain the science behind closing churches but not target.
Greg: that is why polls over the past 10-15 years are not accurate... the demographic needs to be re adjusted by the pollsters... the number of people polled needs to be increased... but 75% is not an accurate and true statistic ...
Skillet I’m not arguing your educational expertise and poll knowledge but polls have been proven to be inaccurate more so in this age than ever before... polls are also showing a preponderance to be technically flawed... many are purposely geared toward swaying opinions and political goals especially now and you state this above... The 75% figure is drastically in error... when polls say “ 75% of the fans say shutdown football“ ... what specific areas of the USA 2200 how many were from the south , southwest, north , northeast and west... the number of polled needs to be a greater number than 2200...
Absolutely embarrassing and pathetic by the Big Ten.
Cannot spin it any other way.
Let’s go Austin Peay.
Cannot spin it any other way.
Let’s go Austin Peay.
That's not a poll that is a prediction. Fitting u wouldn't know the difference. And CNN isn't inovolved with this poll.
Greg : Yes polls are accurate if they properly reflect a proportion across the board of voters or opinions. The point is the 2200 figure is inadequate to formulate a conclusive model of the football fans across the USA. I know Skillet and those statistic gurus will claim mathematically it is etched in stone but the numbers can be easily skewed or misinterpreted by those collecting data. Also a bit of dishonesty is well suspected in most modern day polls. All anyone needs to do is use the 2016 election as a barometer for legitimacy .
the 2016 polls were pretty accurate if you include the MOE. I don't think you have good statistical knowledge about polling based on your posts. It is an established science and 2200 people from across the country is more than enough to get a good poll.
. Greg I understand the MOE ( Margin of Error ) ... I ‘m not as mathematically gifted as some but I still got through college Calculus and Statistics... which I’m happy to say was 52 years ago... Polls can and are manipulated everyday maybe you believe in their legitimacy but many errors have been found ...I maintain the 75% claim is misleading and skewed.
if that poll mentioned earlier has not yet been linked.. here it is
Some highlights...
Only 58% of adults surveyed said they heard a lot or some.. 60% of adults which say they have VERY UNFAVORABLE opinion of Trump... and 8% more with a somewhat unfavorable opinion.. now, of course they could be from the 42% who didn't really hear much about college athletics delays
This poll claims that EVERY ADULT they surveyed that heard some or a lot about the delays AGREES with the delays.. of I suppose it could be that half, 29%, of those that say they heard a lot or some about it think they should play.. but the other 29% from from the people who heard little or nothing about it.
In any case.. the claim that 58% of football fans agree according to this poll is a complete LIE.
The table on page 32 claims of all the people surveyed.. only 4% said they were not a casual or avid fan of college football. So I guess they can claim they are covered that way.. but that statistic seems quite odd. As is the fact they broke out "college football fans" and "avid college football fans".. each with its own column for "avid" and "casual"... and there are more avid college football fans than there are college football fans in that poll.
Poll is highly suspect.
Meanwhile.. maybe the Dem pols are going to want to open things up in order to hope for more cases because as the number of cases drops, Trump's poll numbers on handling the issue gets better.
Some highlights...
Only 58% of adults surveyed said they heard a lot or some.. 60% of adults which say they have VERY UNFAVORABLE opinion of Trump... and 8% more with a somewhat unfavorable opinion.. now, of course they could be from the 42% who didn't really hear much about college athletics delays
How much have you seen, read, or heard about each of the following? College athletic conferences canceling, postponing or rescheduling athletic programs scheduled for the fall, including football
So, the 58% who say they should postpone or cancel the season... that means it matches PERFECTLY the number of adults surveyed who said thaey heard SOME or a lot about the delays.This poll claims that EVERY ADULT they surveyed that heard some or a lot about the delays AGREES with the delays.. of I suppose it could be that half, 29%, of those that say they heard a lot or some about it think they should play.. but the other 29% from from the people who heard little or nothing about it.
In any case.. the claim that 58% of football fans agree according to this poll is a complete LIE.
The table on page 32 claims of all the people surveyed.. only 4% said they were not a casual or avid fan of college football. So I guess they can claim they are covered that way.. but that statistic seems quite odd. As is the fact they broke out "college football fans" and "avid college football fans".. each with its own column for "avid" and "casual"... and there are more avid college football fans than there are college football fans in that poll.
Poll is highly suspect.
Meanwhile.. maybe the Dem pols are going to want to open things up in order to hope for more cases because as the number of cases drops, Trump's poll numbers on handling the issue gets better.
Last edited:
Some of what you say is true, Bob. What has happened in the past 20 years is that polling has flourished as a business, and some pollsters treat it like a business: they cut costs to make money. Hell, in the last election, Daily Kos hired a fire that turned out not to be polling at all. They were just looking at what others were saying, giving it a bit of a left lean, and then making the numbers up! But they got caught at it because their numbers actually looked too logical.
If you read the poll from the polling company, you can find out most of the info you want. But you really have no basis for saying that the poll is drastically in error. It may strongly disagree with what you believe, but Trafalgar has no interest in being dramatically wrong. They are selling that poll to the media to get their name out there. If proven wrong, their business tanks. Most polling companies do their major political polling as a "loss leader" and then they sell their services to marketing groups and businesses.
The sample size was well big enough for what they wanted. Margin of error, which they want to keep reasonable, is a function of sample size and the percentage obtained. So you work backwards from your desired margin of error to see how many folks you need. All that stuff is engineering, and unless someone is deliberately cheating, it's pretty much straightforward.
Here's where you need to keep an eye on pollsters: how are the questions worded and when was the poll taken. Rasmussen, a conservative poll, plays with this sort of thing all the time. Right now you will see them way out of line with the other major pollsters. They are trying to push a narrative. As we get close to the election, you will see them come back in toward what the other major pollsters are saying so that they can claim that their final poll was accurate. Realclearpolitics does a similar thing with their "poll of polls" average. I've watched this every day for years. Since RCP has a right lean, polls which help Trump (and Bush before him) are left in the average longer, and those against him are taken out sooner. They don't do this dramatically, because they'd get caught, but they do it enough to sway the poll 1-2% consistently.
On the other side of the equation, fivethirtyeight.com uses elaborate weighting factors in their polling average. I think they do it honestly, but they generally lean left, so if I were a conservative, I'd be leary. But they called the national level election almost spot on for the past several elections, and had given Trump a decent chance at winning in 2016, when almost nobody else did.
Fun stuff if you are a bit of a math geek.
And listen, I'd love nothing more than for there to be a reverse of the decision and a full football season, but I side with the folks who think it's probably too big a risk. But I certainly understand the sentiments of those who want to play.
Sorry for rambling on, but I'm a professor and I'm professionally obligated to do so.
In many churches, there is a high level of proximity over an extended period of time. There is also singing, in some churches shouting, and often a good amount of physical contact. Unless things have changed since I left, people almost never sing in groups at Target.
If I may offer my $.02 about 2016 polling. The nationwide polls were actually quite close to being correct. IIRC the national average was Clinton +4, and the actual results were Clinton +2. There was some polling error but not all that much.
As Skillethead stated, Fivethirtyeight.com was the only polling aggregator who gave Trump any real chance (about 30%) to win. All the others had Clinton as a near-lock. Why? Because fivethirtyeight.com was the only one to acknowledge that polling error in states could be interrelated.
All the polls had Clinton winning PA, MI, and WI. Most aggregation algorithms were based on the erroneous logic that each state's polling accuracy was independent. While polling might be off in one state, it would be highly unlikely that the polling would be off in multiple states. Fivethirtyeight correctly concluded that if the polling was off in one state, the same factors which caused the polling error could be present elsewhere. Which is what happened.
As Skillethead stated, Fivethirtyeight.com was the only polling aggregator who gave Trump any real chance (about 30%) to win. All the others had Clinton as a near-lock. Why? Because fivethirtyeight.com was the only one to acknowledge that polling error in states could be interrelated.
All the polls had Clinton winning PA, MI, and WI. Most aggregation algorithms were based on the erroneous logic that each state's polling accuracy was independent. While polling might be off in one state, it would be highly unlikely that the polling would be off in multiple states. Fivethirtyeight correctly concluded that if the polling was off in one state, the same factors which caused the polling error could be present elsewhere. Which is what happened.
All these arguments about polling being correct in 2016 ignore the fact that the polls were all wrong in one direction. If this was just an margin of error or another technical issue some would have predicted a much larger Trump victory than actually occurred.
That's why nobody believes them. They are all BS.
That's why nobody believes them. They are all BS.
"The real problem would have to do with non-response rates and bias resulting from them. Unfortunately, most pollsters no longer report that."
Honest question: How accurate can polling be if 77% of Repubs are self-censoring poll questions?
A new Cato national survey finds that self‐censorship is on the rise in the United States. Nearly two-thirds—62%—of Americans say the political climate these days prevents them from saying things they believe because others might find them offensive. The share of Americans who self‐censor has risen several points since 2017 when 58% of Americans agreed with this statement.
These fears cross partisan lines. Majorities of Democrats (52%), independents (59%) and Republicans (77%) all agree they have political opinions they are afraid to share.
https://www.cato.org/publications/survey-reports/poll-62-americans-say-they-have-political-views-theyre-afraid-share
Last edited:
Cato being an arch right-wing source, my attenae are up on their poll (just as they would be on something from DailyKos or the like). So as Warner would say, "I go to the videotape." Cato didn't ask people how they would respond to a poll. That's actually unfortunate, because it would be a cool question. What they asked (stated and asked people to agree/disagree with) is whether the political climate prevents them from saying things that others might find offensive. That's pretty different from responding to a pollster. Note that Cato calls 62% "nearly two-thirds" when it is closer to three fifths, and that they highlight an increase of 4%, which would be statistically insignifcant given the sample size. That's the political twist you see coming from both the left and the right.
Your underlying question is a really good one and one that bothered pollsters a lot in the 2016 election. But despite all the hype, fivethirtyeight.coms final estimate of the election results based on their weighted average of polls was almost spot on. The state polls, often run by lesser quality firms, had a number of errors in them, however.
All that said, I would still be concerned about bias due to moderates not saying they would vote for Trump. I think there might be something there, but I don't have any data on it.
EDIT: And one other thing! "Non-response bias" means refusal to participate (I'm guessing you knew that but others might have it confused with not wanting to say you were voting for Trump and saying you'd vote for Biden instead.).
Cato is actually considered a libertarian think tank, not arch right wing. But we can debate that if the CE Board ever returns lol. Anyway, it seems to me that if 77% of R's aren't comfortable sharing their views in general then it's fairly safe to assume that would carry over to polling questions and therefore be material in poll results, whether the R's are simply not participating or not answering candidly.