Ipsos Libs 38 PC 36 ....
- Thread starter nottyboi
- Start date
Especially the ones from China who don't speak any English or French, who are given the answers to the citizenship test in advance.
It was a completely valid and relevant question.
Without any comparisons, your declaration that Ipsos got 17% is absolutely meaningless.
I'm not sure why my question bothered you so. The only possible assumption would be that Ipsos did better than the other polling companies. Cool.
If YOU want to know the answer, YOU can do the research.
Do it! Do it now! Schnell!
I don't, and I won't.
The 17% accuracy of THAT Ipsos poll was sufficiently noteworthy that it was posted on wikipedia. The comparison of several public opinion polls from years ago is irrelevant to me.
I also don't care about the accuracy of the weather forecast for October 19, 1983.
The obvious conclusion as to why you are so reluctant to post the results of other polling companies is that the results clearly contradict the point you were trying to make. As such, I have no need to do any of my own research. You've provided me with the information I was asking about. Thank you.
If the other polls had 50% accuracy you'd be posting them in bold and enlarged font instead of your very defensive response. The comparison to other polls may be irrelevant to you but they are extremely relevant to people who want to rationally know what your number in a vacuum means. You are obviously trying to denigrate Ipsos' validity but without us knowing how they compare to other pollsters, you have failed.
This sounds like you're giving props to Ipsos for their 17% accuracy in results. That's not unlike a potato chip company boasting that their newly improved product causes 23% less anal leakage than the previous version.
What I posted:
"From wikipedia, re Ipsos:
Polling accuracy
During the 2021–2022 United States election cycle, in the last 21 days before each election, Ipsos polls only correctly predicted winners 17% of the time, according to a FiveThirtyEight analysis of elections for state governor, U.S. Senate, and House of Representatives."
*******************************
To obtain the information YOU wanted, all YOU had to do was go to wikipedia, search for Ipsos, scroll to the polling accuracy paragraph, then click on link #29, which would bring you to this page:
The Polls Were Historically Accurate In 2022
Let’s give a big round of applause to the pollsters. Measuring public opinion is, in many ways, harder than ever — and yet, the polling industry just had one of…
fivethirtyeight.com
At that point, you could have scrolled to the paragraph titled
The most and least accurate pollsters of 2021-22
and seen for yourself that, of the 33 pollsters listed, Ipsos was more accurate than only two others.
The obvious conclusion is that your foot is now in your mouth, again.
Thank you for my early evening dose of schadenfreude.
Not in the slightest. This is you being defensive again. I made no claims regarding Ipsos' efficiency. All I did was ask how they performed relative to others. Somehow this innocent question was a problem for you. That's on you, not me.
Not fair. He knows a few things, but definitely not nearly as much as he thinks he does.
A monkey slinging his poop at photographs of the candidates will correctly predict Senate, House and gubernatorial general elections at least 17% of the time within a 3.4% error percentage, nineteen times out of 20.
The fivethirtyeight chart in the link from post #46 is based on data since 1998. Ipsos hasn't been releasing US State election polls for all of that time, but a 17% success rate says to me that one ought to be skeptical about their skills in prognostication.
When I administered a pension plan for Teamsters some years ago, I was once speaking to a member from close to the Michigan border. He suggested that it might be a good idea for him to delay his unreduced early pension by a month, to increase the amount. I told him that the amount would increase by two dollars and three cents per month, but it would take him seventeen and a half years for that increase to be larger than the initial cheque. He said "Oh. I think limited, eh?".
Last edited:
