Possible Flaws in Rasmussen Poll: Is the Gingrich Surge Real?

Share Button

Newt Gingrich surges in latest set of polls: Image by Gage Skidmore

A Rasmussen poll released yesterday has former house speaker Newt Gingrich maintaining a substantial lead over all other contenders for the GOP nomination. With 38 percent favorability, Gingrich leads Mitt Romney by 21 points. All other candidates trail with single digit ratings.

The GOP race has thus far been peppered with many twists. Leaders in the polls have included Michelle Bachmann, Mitt Romney and Herman Cain. Over the summer, Gingrich’s campaign was on the verge of implosion due to debt and conflict among advisors, and many political observers predicted that he might drop out of the race. Yet, the Iowa caucuses are nearly a month away, and Gingrich has quickly secured a healthy lead, according to Rasmussen.

Rasmussen GOP Nomination Poll Methodology

The Rasmussen survey was conducted on November 30 via automated phone interviews.  Question order was randomized, as was phone number dialing.  The sample consisted of 1000 likely Republican voters who were contacted between 5 P.M. and 9 P.M. local time.  Likely voters were identified by stated willingness to vote in the GOP primary, and were weighted for demographics.  Rasmussen’s general methodology description also states that it uses “a dynamic weighting system that takes into account the state’s voting history, national trends, and recent polling in a particular state or geographic area” to determine partisan weighting targets. So, it is possible that stated party affiliation may have been a weighting factor, even though the entire sample was self-identified Republicans. Opinions of other self-identified Republicans in different parts of the state or nation may have been used to weight the sample.The margin of error was +/-3 points.

Rasmussen Poll: Flaws

Newt Gingrich poll numbers are rising. Image by Decoded Science

Rasmussen has previously been the subject of criticism for inaccuracy and bias, particularly by polling analyst Nate Silver of FiveThirtyEight.com. Although dissenters have typically accused Rasmussen of favoring the right over the left in its polling, this particular poll indicates there may be either bias or methodology flaws in the way Rasmussen gathered opinions regarding the 2012 Republican primary. Sample size in this poll was relatively small, but a bigger concern is that the entire poll was apparently conducted over the course of four to six hours. Scientifically-accurate polls are generally conducted over several days, with respondents being contacted throughout different times of the day, to ensure a representative sample. Automated surveying also poses a problem, since many respondents may hang up, or become confused while trying to navigate a recorded system. Rasmussen’s methodology statement is vague in its description of party-based weighting methods, and these may have affected the polling process if they were used unnecessarily as well.

Presidential Nomination: Interpreting the Poll Results

Aggregate data from RealClearPolitics.com affirms that Gingrich is in the lead for the GOP Presidential Nomination. However, the margin of his lead averages out to 6 points, based on five recent polls, including this particular Rasmussen poll. The 21 point lead reported by Rasmussen jumps out as being exceedingly large, when compared to the other polls, in which Gingrich’s lead ranges between one and four points. While it is reasonable to conclude that Gingrich has, in fact, pulled ahead in the race, the monumentus leap indicated by this particular poll is not substantiated by aggregate data.

Sources

Rasmussen Reports. Election 2012: Republican Presidential Primary. Accessed December 2, 2011.

Rasmussen Reports. Methodology. Accessed December 2, 2011.

Silver, Nate. Rasmussen Polls Were Biased and Inaccurate; Quinnipiac, Survey USA Performed Strongly. FiveThirtyEight.com. Accessed December 2, 2011.

RealClearPolitics. 2012 Republican Presidential Nomination. Accessed December 2, 2011.

Share Button
© Copyright 2011 Marissa Selner, All rights Reserved. Written For: Decoded Science
Decoded Everything is a non-profit corporation, dependent on donations from readers like you. Donate now, and keep the great information coming!

Trackbacks

  1. [...] from standard good polling practices has made Rasmussen’s methodology controversial among polling analysts. Rasmussen touts its [...]

  2. What I’m Reading Friday, December 2, 2011 | Rationally Thinking Out Loud says:

    [...] Flaws in Rasmussen Methodology: Is Gingrich Really in the Lead? | Decoded Science [...]

Speak Your Mind

*