How the Question is worded effects the Poll

Damocles

Damocles

Accedo!
Staff member
http://www.rasmussenreports.com/pub...vember_2009/question_wording_and_job_approval

Monday, November 30, 2009

Not surprisingly, when you ask about the President’s Job Approval in different ways, you get different results.

On Tuesday, November 24, 2009, Rasmussen Reports asked three separate samples of 800 Likely Voters a question about the President’s Job Approval. In all three cases, it was the first question asked on the survey. Survey, sampling, and weighting techniques were identical for all three. See toplines and question wording.

When we asked the question in the standard Rasmussen Reports format, we found that 47% of voters approved while 52% disapproved. Those figures include 28% who Strongly Approve and 41% who Strongly Disapprove.

However, we got different results when we simply asked if people Approved or Disapproved. When we took away the options of Strongly Approve/Disapprove and Somewhat Approve/Disapprove, the results were 50% approve and 46% disapprove. It’s important to note that the difference could be just statistical noise in a survey with a +/- 4 percentage point margin of sampling error.

(Want a free daily e-mail update? If it's in the news, it's in our polls). Rasmussen Reports updates are also available on Twitter or Facebook.

However, the difference is consistent with years of observations that Rasmussen Reports polling consistently shows a higher level of disapproval for the President than other polls. That may reflect the fact that there may be some people willing to offer a “somewhat disapprove” rating rather than say they “disapprove.”

More at link...
 
The number of options and the wording both affect the outcome of any given poll. For example:

Would you rate the presidency of Barack Obama over the reign of Adolph Hitler?

That question would obviously bring a high percentage in his favor. Hyperbole yes, but releventg as well.
 
CaptBillyTheKid said:
The number of options and the wording both affect the outcome of any given poll. For example:

Would you rate the presidency of Barack Obama over the reign of Adolph Hitler?

That question would obviously bring a high percentage in his favor. Hyperbole yes, but releventg as well.
Click to expand...

I wasn't saying it didn't. It's just not the point of this article.
 
Watermark said:
I wasn't saying it didn't. It's just not the point of this article.
Click to expand...

Here is an extract from a very well known American book. Here it talks of the American consumer (also read British/any-ish and consumer of products and politics). The manipulation of questionaires and research vehicles is nothing new.

He or she is dramatized as a thoughtful voter, rugged individualist,
and, above all, as a careful, hardheaded consumer of the wondrous
products of American enterprise. He is, in short, the flowering of
twentieth-century progress and enlightenment.
Most of us like to fit ourselves into this picture, and some of
us surely are justified in doing so. The men and women who hold
up these glowing images, particularly the professional persuaders,
typically do so, however, with tongue in cheek. The way these persuaders—
who often refer to themselves good-naturedly as "symbol
manipulators"—see us in the quiet of their interoffice memos, trade
journals, and shop talk is frequently far less flattering, if more interesting.
Typically they see us as bundles of daydreams, misty hidden
yearnings, guilt complexes, irrational emotional blockages. We are
image lovers given to impulsive and compulsive acts.

If it was possible I would award two prizes. The first to the first person to correctly identify the year of publication and the second to the person who can correctly identify the author.
 
You must log in or register to reply here.
Back
Top