Keep the Monsters Away
Donate, Monthly If You Can
Sponsoring FReepers leapfrog0202 and others will contribute $10
Each time a new monthly donor signs up!
Get more bang for your buck
Sign up today
Save Our Sweet (SOS) Lazamataz
Posted on 05/09/2011 11:14:58 AM PDT by MattAMatt
An article in the Washington Post, entitled, "Bin Laden death helps Obama reposition in Virginia" attempts to calculate the impact of Osama Bin Laden's assassination on Virginian's perception of Obama. A random cell/home phone survey was conducted before and after Obama's announcement on Sunday, May 1st. Respondents were asked 34 questions that related to both Virginia State politics and National races. The polling data showed that Obama's approval rating increased by 8% when compared to results of the same survey conducted before the assassination. They also showed that Obama stretched his lead against the top five potential Republican Candidates. Unfortunately, after reviewing the complete poll data it was clear that the results were likely to be invalid.
Inadequate Disclosure Full disclosure is the first sign that research is done well. It is the standard practice of the polling profession to release the full questionnaire, the results and the methodology. Most people begin to doubt the accuracy of polling data after they see "Question 1 held for release." Doubts easily turn into certainty after finding that 10 questions in addition to the first one are also "held for release." Even The New York Times acknowledges that its the standard practice of the polling profession to release the full questionnaire, the complete results and the methodology.
Methodology The Washington Post not only fails to meet the minimum standards of disclosure by omitting questions/results, but also by inadequately disclosing their methodology. According to The American Association for Public Opinion Research (AAPOR), it is unacceptable to release surveys without also disclosing certain kinds of information about how the poll was conducted. The incomplete polling data they disclosed only included half of the essential information needed to show how the research was conducted. Two excerpts from the poll data have been pasted below followed by a list of questions that, when answered, could change the results and/or expose inaccuracy/bias.
According to The Washington Post, the poll was conducted by telephone April 28 to May 4, 2011, among a random sample of 1,180 adults in the Commonwealth of Virginia, including 1,040 registered voters and users of both conventional and cellular phones.
1. We still need to know what the sample selection procedure were?
2. Were the respondents selected by the researcher,or were they voluntary, self selected?
3. What was the eligibility criteria and screening procedures?
4. Were the respondents contacted in a specific order i.e. by zip,street etc.?
5. Did 100% of the 1180 adults agree to do the survey?
6. How many home/cellular phones were called to reach 1180 respondents?
7. What was the completion rate? What was the ratio of Cell to Home Phones?
8. Was it possible to prevent/determine if their were two respondents in one home?
"677 interviews conducted before targeting killing of Osama bin Laden; 503 afterward. The results from the full survey have a margin of sampling error of plus or minus 3.5 percentage points. Sampling, data collection and tabulation by Abt-SRBI of New York"
1. How many phone numbers were contacted in total?
2. Were there any numbers contacted both before and after due to no answer/not home?
3. How do you tell if you are surveying the right person?
4. What shows up on the caller ID of the respondent?
5. Why was their a smaller sample size after the assassination?
6. Is there a demographic breakdown per question (before/after) i.e. age,sex,marital status, income etc.?
7. What was the ratio of Democrats to Republicans? Independents? Libertarians?
Timing Impacts Responses Current events always sway the results of polls in one direction or the other. The attacks on 9/11 is one example of an event that will have a long term impact on opinion polls. If a question about the threat of terrorism was asked a day before 9/11 and then again the day after, the results would be polar opposites. However, the answer today would be similar to post 9/11, but be trending in another direction. The Washington Post survey was conducted the day after Osama's assassination and results are representative of a snap shot in time, not a trend. It was completed well before Americans had an opportunity to digest the reality and their emotions so they could form opinions.
The campaign season: A period in time when suspense and hope are propulsory forces that cultivate ardent support and inspire superhuman efforts. What normally takes a month to accomplish in an office, can be done in one day on the campaign trail. A period when time seams to stand still and then, In the blink of an eye, momentum changes, front runners fall and the impossible becomes possible. Seasoned political supporters understand this, and are far too stubborn and/or competitive to be swayed by questionable newspaper polls. They know there is always room for hope, even if poll results show that it's 90%-3% with 7% undecided. Most importantly, campaign supporters know that poll results are easily manipulated by slanted questions, inadequate sampling, inferior methods, incomplete poll data and/or misrepresentation of results.
Okay, so now the MSM is calling Bin Laden’s death an assassination?
So it wasn’t an action as part of a war, a war declared by Bin Laden on us, it was an “assassination”?????
But, the liberals have to reconcile the fact that this “assassination” was authorized by a good liberal president.
I really wonder what the reaction would have been if this happened to Bin Laden while Bush was president.
I’m calling it an assassination and am trying to use Key word density to get SEO ranking, so more people see it as an assassination. It’s time this President was held accountable.
And what about the bump on Rasmussen?
What about all of the Bumps he is getting now. Fact is these increases are not trends. Therefore, any minor event or gaff can bring it right back down.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.