Search This Blog

Showing posts with label polls. Show all posts
Showing posts with label polls. Show all posts

Sunday, November 13, 2016

A Conventional Election?

There are at least three ways in which this very unconventional election was surprisingly conventional.

First, the polls said that Clinton would win the popular vote.  She did.

Second, the poli sci models were not far off the mark.  John Sides writes at The Washington Post:
In January 2014, I wrote a piece called “The Democrats’ uphill battle to 270 electoral votes.” That piece was predicated on conditions in the country at the time: economic growth, the president’s approval rating and the greater tendency of the White House to change parties after two terms than one term, also knows as “time for a change.”

In March 2016, Michael Tesler, Lynn Vavreck and I drafted an article — subsequently published in August 2016 — in which we suggested that there were mixed signals in the American electorate. People’s views of the economy had recovered, although a variety of political assessments remained more pessimistic. Taken together, economic growth and presidential approval gave the Democrats only a modest chance of winning. If you factored in “time for a change,” the Republicans were favored.
Other early forecasts by political scientists and economist Ray Fair showed the same thing. Vox took several such forecasts, asked two political scientists to average them, and came up with a narrow Republican victory in the national popular vote. Yesterday, the full roundup of these forecasts by Pollyvote said that Clinton would win the popular vote by 0.4 points. Currently, she has a 0.2 percent lead. She will probably exceed that, but the point stands: Those models said it was supposed to be close.
Then, like most observers — including, apparently, the Trump campaign — I believed the forecasting consensus that Hillary Clinton would win and wrote as much yesterday. There were forecasts that gave Trump at least a reasonable shot, including the 24 percent chance given by the forecasters at our partner, Good Judgment. But I went with that consensus and, like others, thought Trump would underperform the average of the political science forecasts. So I was wrong too.
Much will be written about why the polls in particular underestimated Trump’s vote share. It is all the more puzzling because serious attempts to measure the “hidden” Trump vote struggled to find much evidence of them.
But for now, a striking fact is that the election ended up looking a lot like an average of the fundamentals-based models from several months ago.
Third, Todd Eberly writes that the electoral vote displayed a surprising degree of continuity.
... I paired successive elections by comparing the Democratic share of the total vote by state to generate correlation coefficients. The state-by-state comparison generates a Pearson correlation coefficient with a low or negative value indicating a disruption or cleavage point and a higher positive value indicating continuity or electoral stability.  Figure 1 presents the results. The changes taking place in the parties' coalitions in the late 1970s and early 1980s are evident as is the rather stable electoral period that emerged in 1996 - a period marked more narrow election victories and the emergence of the familiar Red and Blue states.
Donald Trump pieced together an impressive electoral coalition in the 2016 election, a coalition that included traditionally Democratic states like PA, MI, and WI - yet Figure 1 shows the 2016 election demonstrated a high degree of correlation with the election of 2012 – no fundamental changes took place in 2016. Rather Trump was able to achieve narrow victories in otherwise Democratic states. The results suggest that the present path to 270 electoral votes remains both stable and competitive.

Monday, August 24, 2015

Billionaire Surges in Polls!

For some perspective on billionaire presidential candidates surging in the polls, consider The New York Times on June 11, 1992:
In a three-way general election matchup, Ross Perot has moved to a clear lead over both President Bush and Gov. Bill Clinton in the latest Gallup Poll.
In the telephone poll of 815 registered voters nationwide, conducted June 4 to 8, Mr. Perot was supported by 39 percent, Mr. Bush by 31 percent, and Mr. Clinton by 25 percent. The poll had a margin of sampling error of plus or minus four percentage points.
In a previous Gallup matchup in late May, Mr. Bush and Mr. Perot were tied at 35 percent each, while Mr. Clinton was supported by 25 percent.
No previous independent or third party candidate has ever placed second, much less first, in nearly six decades of Gallup's nationwide polling for President.
Another poll found a double-digit Perot lead,  37 percent to 24 percent each for Clinton and Bush.

On May 28 of the same year, USA Today reported:
Anyone doubting Ross Perot is for real can look at the polls, where he is showing electoral strength impossible to ignore. The evidence is in 25 statewide polls, representing more than 60% of the Electoral College votes:
- Perot leads in nine states having 128 electoral votes.
- President Bush leads in 15 states (190 votes). But much of that came in early polls, before Perot ran first in a Texas poll April 18. Since then he's won seven state polls to Bush's 10.
- Democrat Bill Clinton has led only at home in Arkansas, worth six electoral votes. It takes 270 to win.
Perot ended with 18.9 percent of the popular vote and zero electoral votes.

Sunday, November 16, 2014

Missed GOP Opportunities

It was a good election for the GOP but it could have been even better.

Steven Shepard writes at Politico:
For two years, Republicans had been working to correct one of the party’s greatest embarrassments of recent years: the flawed polling that led so many in the party to believe Mitt Romney was on the cusp of victory in 2012. But after dramatically underestimating Democratic turnout in 2012, it was now obvious that the GOP had erred in the other direction in 2014. Their pollsters had understated Republicans’ leads in a number of states, causing the RNC and GOP campaign committees to pour money into places where it wasn’t needed and hold money back from places where it might have made a difference — such as Virginia, where Republican Ed Gillespie lost by less than a percentage point.
“It’s just as bad to be wrong by being too conservative,” said [RNC chief of staff Mike] Shields. “It’s just as big a mistake to tell a client that you’re only winning by one point when they’re winning by eight. Especially at the party committee level, there are just too many decisions being made … That money can be used elsewhere.”
Jon Fleischman writes at Breitbart:
Nationally, Republicans had an outstanding election earlier this month. Besides the huge gains in the U.S. Senate, House Republicans made modest gains (12 seats), bringing their majority to an impressive 244 of 435 members. That said, one cannot help but note that House Republicans could have had an even better night but for being completely routed by Democrats in California. Here is a snapshot of nine separate House races, all likely won by Democrats, that could have gone Republican.
...
Obviously hindsight is, as they say, 20-20. But there is simply no way to look at all of these results and not conclude that, had the NRCC made different decisions, there could have been a whole bunch of new freshman Republicans from California, replacing Democrats. 2016, a Presidential election year, will make all of these seats more difficult for GOP pickups in two years.
It is worthy of note that strong performances were made by two GOP incumbents in tough seats in California — both in the Central Valley — Jeff Denham and David Valadeo. And it should be said that this year the California GOP turned the tide back a bit in the state legislature, going from 12 up to 14 members in the 40-member State Senate, and from 25 to 28 members in the 80-member State Assembly. These state legislative gains were in no small part due to a strong focused and well-financed plan largely orchestrated by State GOP Chairman Jim Brulte.

Tuesday, August 30, 2011

It's Still Early

At The Chicago Tribune, Eric Zorn offers a useful note of caution about early polls:

A Fox News/Opinion Dynamics poll taken in late August 2007 — nearly this exact point in the previous election cycle — had former New York City Mayor Rudy Giuliani way out in front of his Republican rivals, and Arizona Sen. John McCain, who eventually won the nomination, limping along in fourth place with 7 percent support.

Among Democrats in that same survey, New York Sen. Hillary Rodham Clinton was pounding the eventual winner, Illinois Sen. Obama, 38 percent to 25 percent. In early September 2007, CNN/Opinion Research had Clinton up 48 percent to 23 percent, and McCain still lagging well behind Giuliani and former Tennessee Sen. Fred Thompson.

These polls weren't outliers. Giuliani and Clinton led in every major scientific survey from that time archived at the indispensable PollingReport.com. McCain didn't surge into the lead in any national poll until the second week of January 2008. Obama didn't pass Clinton until the first week of February.

Eight years ago at this point in the cycle, a pack of Democrats was running to challenge Republican President George W. Bush. A Fox poll had Connecticut Sen. Joe Lieberman in the lead, as did surveys by CNN/USA Today/Gallup and Quinnipiac University. Newsweek had retired U.S. Army Gen. Wesley Clark in front, NBC News/Wall Street Journal concluded former Vermont Gov. Howard Dean was the narrow favorite of his party, and only Time/CNN found the eventual nominee, Massachusetts Sen. John Kerry, in first place.

Gallup, however ,has noted that early Republican frontrunners do tend to go on to the nomination.


Tuesday, March 31, 2009

New Research on 2008 Nomination Politics

A new report from the American Association for Public Opinion Research examines the polls that falsely indicated that Barack Obama would win the New Hampshire primary. It concludes:
  • Given the compressed caucus and primary calendar, polls conducted before the New Hampshire primary may have ended too early to capture late shifts in the electorate’s preferences there.
  • Most commercial polling firms conducted interviews on the first or second call, but respondents who required more effort to contact were more likely to support Senator Clinton. Instead of continuing to call their initial samples to reach these hard‐to‐contact people, pollsters typically added new households to the sample, skewing the results toward the opinions of those who were easy to reach on the phone, and who more typically supported Senator Obama.
  • Non‐response patterns, identified by comparing characteristics of the pre‐election samples with the exit poll samples, suggest that some groups who supported Senator Clinton—uch as union members and those with less education—ere under‐ represented in pre‐election polls, possibly because they were more difficult to reach.
  • Variations in likely voter models could explain some of the estimation problems in individual polls. Application of the Gallup likely voter model, for example, produced a larger error than was present in the unadjusted data.
A new experimental study finds:
After Clinton's concession in June 2008, Democrats were more generous toward supporters of their own preferred candidate than to supporters of the other Democratic candidate. The bias observed in June persisted into August, and disappeared only in early September after the Democratic National Convention. We also observe a strong gender effect, with bias both appearing and subsiding among men only
.