Posted on 11/05/2018 9:57:14 AM PST by SpeedyInTexas
You’ve heard of the “margin of error” in polling. Just about every article on a new poll dutifully notes that the margin of error due to sampling is plus or minus three or four percentage points.
But in truth, the “margin of sampling error” – basically, the chance that polling different people would have produced a different result – doesn't even come close to capturing the potential for error in surveys.
Polling results rely as much on the judgments of pollsters as on the science of survey methodology. Two good pollsters, both looking at the same underlying data, could come up with two very different results.
(Excerpt) Read more at nytimes.com ...
Results varied by 5 points between pollsters reviewing the same raw data.
Tomorrow some pollsters will be eating crow. And crow "is a dish best served cold".
I’m sure you remember this article.
I gave four Italian pollsters the same data and got five different opinions.
I know that’s more of a Jewish joke but I’d have to convert before I could use it.
I gave four Freepers the same post and got four different replies.
May they choke on the feathers.
Do not hang your hat on polls
Actually a good article for Pravda West
Excellent!
Yep definitely do...
...and Trump won Florida by 1.2%
Amen. The game ain’t over. Vote!
sounds like hurricane computers.
Feed the same raw data into 31 computers and they spit out 31 different hurricane tracks. And we are supposed to have confidence in computers to make life and death decisions??
This is true if you believe they don’t manipulate the data or flat out manufacture the numbers, which I do not. What assurance is there, other than “their reputation”? Gimme a break, the only people who care about polls are those paying for them. And they are not interested in accuracy or objectivity.
Are you planing to vote?
Who for?
Thank you for your time.
Call 2000 people per state.
You will get the best and most truthful data.
Analysts is all about the assumption.
Analysis, rather.
How many Poles does it take to screw up a poll?
ALL “scientific” forecasting (including weather/climate modeling, election or economic predictions, etc) includes unverifiable assumptions by the “scientists.” These assumptions include voter turnout predictions, consumer buying predictions, effect of CO2 on climate, etc. These assumptions often can be influenced by political ideology, availability of grant money, and other factors. That is why there have been 14 different climate change models and NONE of them have accurately predicted the weather during the past decade.
The secret to selling $ucce$$ful polling to a customer is $imple. Find out what an$wer he wants and give it to him. If it turns out wrong blame the respondents or something else. Promise to do better next time.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.