Thursday, October 6, 2016

Latest polls from HuffPollster

Clinton leads in New Mexico (bigly) and Ohio (barely) - from HuffPollster email.


-New Mexico: SurveyUSA and KOB-TV find Hillary Clinton leading Donald 46-33; Pollster average gives her a 10 point advantage. [SurveyUSA, chart]

-Ohio: Monmouth finds Clinton up 44-42, and GOP Senate incumbent Rob Portman leading 54-39; Quinnipiac has Portman ahead 55-38;Pollster average shows a less than 1-point margin in the presidential race, and Portman up by 13. [Monmouth, Quinnipiac, presidential chart, Senate chart]

But take note: polls are less accurate than the statistics show.

POLLING ERROR HAS TO DO WITH MORE THAN JUST SAMPLING - David Rothschild and Sharad Goel: “As anyone who follows election polling can tell you, when you survey 1,000 people, the margin of error is plus or minus three percentage points. This roughly means that 95 percent of the time, the survey estimate should be within three percentage points of the true answer….But the real-world margin of error of election polls is not three percentage points. It is about twice as big. In a new paper with Andrew Gelman and Houshmand Shirani-Mehr, we examined 4,221 late-campaign polls — every public poll we could find — for 608 state-level presidential, Senate and governor’s races between 1998 and 2014. Comparing those polls’ results with actual electoral results, we find the historical margin of error is plus or minus six to seven percentage points….All these nonsampling errors show up in two ways. First, polls within a race vary from one another slightly more than one would expect from classical textbook explanations. Second, and most markedly, polls tend to systematically overestimate or underestimate the true answer….This November, we would not be at all surprised to see Mrs. Clinton or Mr. Trump beat the state-by-state polling averages by about two percentage points. We just don’t know which one would do it.” [NYT]

No comments:

Post a Comment