In the months leading up to the 2012 presidential election, virtually every major poll has had President Barack Obama leading over Republican challenger Mitt Romney.
However, critics contend the polls have a built-in bias intended to boost Obama’s numbers.
At issue is the number of Democrats sampled for polls compared to Republicans. In many major polls, a significantly higher number of self-reported Democrats are surveyed.
Some Republicans say this skews the polls in favor of Obama while those supporting the polling methodology argue it simply reflects the reality of a larger Democratic voting base.
‘Skewed’ Polls Cause Concern
A review of polls by major firms reveals that virtually all survey more Democrats than Republicans. Almost all of these polls also show Obama leading by slim margins; margins conservatives say wouldn’t exist if pollsters surveyed equal numbers of Republicans and Democrats.
Take, for example, the latest United Technologies/National Journal Congressional Connection Poll. Among likely voters, the poll found the Obama/Biden ticket gets 50 percent of the vote compared to 45 percent for Romney/Ryan.
However, Democrats also outnumbered Republicans in terms of those surveyed. Of the likely voters polled for the Congressional Connection Poll, 36 percent identified themselves as Democrats compared to 28 percent who said they were Republicans.
The Congressional Connection Poll isn’t alone in this practice. According to UnskewedPolls.com, a poll tracking site operated by conservative QStarNews , virtually all recent major polls have used models that survey larger percentages of Democrats:
- ABC News/Washington Post: +7 D
- IBD/TIPP: +7 D
- CBS News/NY Times: +5 D
- Rasmussen: +3 D
Presidential Poll Weighting
Those supporting the polls say the numbers simply reflect the fact more Democrats are expected to vote in the upcoming election than Republicans. Many point to the 2008 election as support for this assumption. In 2008, exit polling from key swing states showed Democrats turned out in significantly higher numbers. For example, in Ohio, 39 percent of voters were Democrats compared to 31 percent who were Republicans.
In addition, many polls track trends among registered voters, and have found that Democrats and Independents have made significant inroads in increasing their registered numbers in recent years. A 2009 analysis of voter registration statistics found the number of Democrats registering to vote increased significantly compared to Republicans.
The data, compiled by researchers within the United States Election Project at George Mason University, found voter registrations among Democrats rose 10.8 percent from 2004-2008. Meanwhile, Republican registrations increased 0.5 percent during the same time frame.
Some point to this increase in voter registrations as a reason why more Democrats may be polled than Republicans.