AOL News has a new home! The Huffington Post.Click here to visit the new home of Politics Daily!
If you follow the polls, including those often cited in this blog, then Rasmussen Reports is a familiar name to you. Some news organizations will not cite Rasmussen polls because the surveys it conducts are automated (known as IVRs for "interactive voice response") rather than the kind of live telephone interview polls that are used by organizations like Gallup, Pew Research, Quinnipiac University and the major newspapers and television networks.
But Rasmussen attracted a different kind of attention the other day when Politico ran a story saying that Democrats and the liberal blogosphere were accusing the pollster of bias towards Republicans and conservatives, saying the surveys were "at best, the result of a flawed polling model and, at worst, designed to undermine Democratic politicians and the party's national agenda."
Democratic pollster Mark Mellman told Politico, "I think they write their questions in a way that supports a conservative interpretation of the world. In general, they tend to be among the worst polls for Democrats, and they phrase questions in ways that elicit less support for the Democratic point of view."
Putting aside for the moment an analysis of how his questions are worded, Scott Rasmussen says that one reason some of his polls are trending Republican is that he is surveying likely voters, and the dynamic at work now is that conservatives are more energized than Democrats (which other polls have borne out). Other polls survey registered voters, a smaller percentage of whom are likely to vote, or just "adults."
Nate Silver at FiveThirtyEight.com, a highly regarded analyst of polling, also has his problems with the wording of Rasmussen's questions, but says the operation's polling of political horse races "has tended to be quite accurate in the past."
When Rasmussen comes out with a different result from other pollsters, one that favors Republicans, Silver argues that it doesn't represent bias but "a different model of what the 2010 election is going to look like, one which will feature a more conservative electorate . . . and ultimately, these differences of opinion will be tested -- based on what happens next November."
Silver says that Rasmussen is just one of many pollsters and has this advice for consumers of polling and for those who write about it: "If you're running a news organization and you tend to cite Rasmussen's polls disproportionately, it probably means that you are biased -- it does not necessarily mean that Rasmussen is biased."
As for the reliability of automated polling, Pew's Scott Keeter told the Wall Street Journal's "Numbers Guy" in a past interview: "There are some advantages to IVR -- larger sample sizes for the same fixed dollar amount, possible greater respondent honesty -- offset by disadvantages such as ability to control who is the respondent, limitations on length of survey, and possible bias toward people comfortable with interacting with recorded voice questions."
If you follow this blog, you know that we cite a wide range of pollsters and, in our periodic roundups of how President Obama is faring in polls in individual states, we present all the results in one package, so readers can compare.
Poll Watch primarily uses Rasmussen for campaign polls on Senate and governor races, and much less frequently on general issue polls because I share some of the reservations about the wording of questions -- less because of concern over bias, but more because in areas like the health care debate, I think other pollsters do a far better job in their use of language to probe public opinion.
Nate Silver has an update on this debate at his blog.
News From Our Partners
More on Aol
Sites and Services