A Very Brief Comparison of Rasmussen and the Other Polls and Why Rasmussen Has a Slight Edge Methodologically
Abdul Karim Bangura
As the respected EcoPoliticalEcon (http://ecopoliticalecon.com/2012/05/10/rasmussen-gallup-different-poll-results/) points out, the first, and most obvious, difference between the polls is that the others, such as Gallup, use a sample of American adults while the Rasmussen Poll samples likely voters. Likely voters have different and stronger opinions than randomly selected adults.
Rasmussen uses a Random Digit Plus One telephone survey. While it does not include cell phones, it compensates for this by using an online survey tool. Most adult voters with cell phones also have land line phones. Younger voters who only have cell phones, a very small number who actually vote, are captured via the online survey tool.
The other polls' methodology is slightly different. They also use a Random Digit Plus One telephone survey that includes a minimum quota of cell phone respondents.
All of the polls weight their data to better reflect the overall population in terms of race, gender, age, and party, as well as other factors. They also, as a rule, phrase their questions in as unbiased a manner as possible.
The choice of spread for all of the polls is a 95% confidence interval: i.e. there is a 95% chance, or we are 95% confident, that the true value of support is within the stated interval. So for very large samples, the mean of the sampling distribution is represented by a normal or Gaussian distribution, and the associated z score is 1.96 for the interval. All of the polls also use a margin of error of +/- 3.
Thus, comparatively, the Rasmussen Poll has a slight edge over the other polls in determining the actual voters' preference in an election.
No comments:
Post a Comment