FAQ: Table of Contents
- Why are polls showing contradictory results?
- What does the "margin of error" mean?
- How do pollsters select "likely voters?"
- Should pollsters weight by party identification?
- Do undecided voters break for the challenger?
- Are polls missing people with mobile phones?
- Can polls tell who won a debate?
- Questions about Exit Polls
- Other questions
Why are polls showing contradictory results?
In a sense, every post on this site seeks to answer the question of why polls seem contradictory. A good place to start is with the following three posts on the random variation inherent in all sample surveys and what a consumer can do to minimize it (each post has a link at the end connecting to the next post in the series):
Beyond that, much of the variation in political polls comes from the way the pollster defines a "likely voter.” These two questions on the FAQ list deal with two issues that explain much of the systematic variation between polls:
Finally, at the most basic level, political campaign polls typically seek to accomplish three tasks. Click on any of these links to display short excerpts of all of the posts in the category in reverse chronological order (most recent first):
- Contacting a random sample of adults
- Selecting a pool of potential or likely voters
- Interpreting the results
What does the "margin of error" mean?
A good place to start is with the following three posts on the random variation inherent in all sample surveys and what a consumer can do to minimize it (each post has a link at the end connecting to the next post in the series):
Also, click on this link to display short excerpts of all of the posts concerning sampling error in reverse chronological order (most recent first).
How do pollsters select "likely voters?"
I have done a lot of posting on this topic. If you want a very thorough reading, start with Part I and continue through the series. You might also want to skip to Part VIII, which is something of a guide to the likely voter models used by all the major polling organizations:
- The Why & How of Likely Voters - Part I
- The Why & How of Likely Voters - Part II
- The Why & How of Likely Voters - Part III
- Likely Voters IV - The Gallup Model
- Likely Voters V: More Gallup
- Likely Voters VI: Still More on Gallup
- Likely Voters VII: CBS/NYT
- Likely Voters VIII: Users Guide to Everybody Else
Also,you can click on this link to display short excerpts of all of the posts concerning likely voters in reverse chronological order (most recent first).
Should pollsters weight by party identification?
The debate over whether pollsters should statistically adjust (or weight) their samples by party identification has been heated during campaign 2004. Here is a list of sequence of posts that covered this topic exhaustively:
- Weighting by Party
- More on Party ID
- Why & How Pollsters Weight, Part I
- Why & How Pollsters Weight, Part II
- Why & How Pollsters Weight, Part III
- Why & How Pollsters Weight, Part IV
It is worth noting that the debate about weighting by party is really a part of a larger debate also raging in 2004 about how pollsters define likely voters – click this link for more information.
Finally, click on this link to display short excerpts of all of the posts concerning weighting by party in reverse chronological order (most recent first).
Do undecided voters break for the challenger?
When pollsters interpret results for a race involving an incumbent, pollsters have typically applied the rule that incumbents rarely get a higher percentage in the election than they receive in polls, and that voters still undecided on the very last poll tend to "break" disproportionately for the challenger.
I wrote quite a bit on the incumbent rule before the 2004 election, but then the results at the presidential level defied the rule. I considered some of the reasons for what appears to be a recent weakening of the incumbent "rule" in this post in August, 2006.
These two posts from 2004 best explained the "incumbent rule:
Also, click on this link to display a listing of all posts concerning the incumbent rule on Pollster.com (in reverse chronological order - most recent first). This link display posts on the incumbent rule from the blog MysteryPollster.
Are polls missing people with mobile phones?
These two posts from 2004 address the issue of how households with cell phones but no "landline" service can affect telephone surveys:
Can polls tell who won a debate?
Start with these two posts on how instant polls and focus group aim to evaluate how candidates faired in a debate:
For more commentary, click on this link, or the link for the "debates" category, for a list of short excerpts of all of the posts on polls aimed at measuring reaction to the presidential debates. They will appear in reverse chronological order (most recent first).
Questions about Exit Polls
Since the election, I have written quite a bit on exit polls. Although I have learned a lot in that process, my Election Day summary of "what you need to know" about exit polls still holds up well and is a good place to start if you are new to this site.
Here is the complete list of questions:
- What is the difference between "partial" and "final" exit polls? Covers which tabulations were "corrected" to match the actual count and which were not.
- Where can I find the "uncorrected," just-before-poll closing exit poll tabulations?
- Where can I find official methodological information about exit polls?
- What is the margin of error for exit polls?
- Were the exit polls really "wrong?"
- Why were exit polls in Germany & Utah so accurate?
- Have the exit polls been wrong before?
- If not random error, why were they "wrong?"
- So if exit polls have so much error, why do them?
- I heard the exit poll estimates for Hispanic voters have been revised -- is that true?
- What is MP's reaction to the paper, "The Unexplained Exit Poll Discrepancy," by Steven Freeman? (This post dicusses his first draft at length, this post has Freeman's response).
More frequently asked questions - click the link for answers:
- Can polls tell who won a debate?
- Why do retrospective questions about the 2000 race show Bush beating Gore?
- Why are there "Democratic pollsters" and "Republican pollsters?" Does their partisanship bias the results?
- Are polls missing voters in the military serving out of the country?
- Why blog about polling?
More to come…
Sen General Election:
2008 POLL DATA
Pres General Election:
All 08 GE Match-ups
All 08 Primaries
Democracy Corps (D)
The Field Poll
ICR - International Communications Research
Mason Dixon Polling and Research
Market Shares Corporation
NBC/Wall Street Journal
New York Times
Opinion Research Corporation
Pew Research Center
Princeton Survey Research Associates International
Public Policy Polling
Quinnipiac University Poll
Selzer & Company
Suffolk University Political Research Center
World Public Opinion
POLL BLOGS AND SITES
The Polling Report
R. Chung's Graphics
Prof. Wang's State Poll Meta-Analysis
Prof. Pollkatz Pool of Polls
Slate: Election Scorecard
Public Opinion Pros
Frank Newport: Gallup Guru
Carl Bialik: The Numbers Guy
Poll Positions: Kathy Frankovic
The Numbers: Gary Langer
Washington Post: Behind the Numbers
SURVEY RESEARCH ORGANIZATIONS
American Association for Public Opinion Research (AAPOR)
The National Council on Public Polls (NCPP)
Council of American Survey Research Organizations (CASRO)
The World Association for Public Opinion Research (WAPOR)
The Council for Marketing and Opinion Research (CMOR)
Marketing Research Association
July 20, 2008 - July 26, 2008
July 13, 2008 - July 19, 2008
July 6, 2008 - July 12, 2008
June 29, 2008 - July 5, 2008
June 22, 2008 - June 28, 2008
June 15, 2008 - June 21, 2008
June 8, 2008 - June 14, 2008
June 1, 2008 - June 7, 2008
May 25, 2008 - May 31, 2008
May 18, 2008 - May 24, 2008
All pollster.com archives