Re: interpreting poll margin of error?

From: Leo G Simonetta (lsimonetta_at_newsguy.com)
Date: 10/25/04 

Date: Mon, 25 Oct 2004 17:07:56 -0400

On Sat, 23 Oct 2004 14:53:30 GMT, Art Kendall <Art@DrKendall.org>
wrote:

> Stan makes a vital point about extra-statistical considerations.
>
> The total uncertainty about the meaning of a statistical result is made
> up of uncertainty due to sampling, and the "big sources" of uncertainty.
>
> The perceived or actual polemic position (often misnamed rhetorical
> position) of the sponsor is one, but since pollsters live in glass
> houses, and because many sponsors want as objective and impartial
> information so they can get to design strategy effectively, often a
> great deal of effort is made to minimize this.
>
> But here are always the questions about the non-statistical uncertainty,
> like those that follow.
> Is the population intended identical to the population about which we
> wish to make rhetorical (in the old sense of the word as principled
> argument) statements? (frame errors -- respondents not identical to
> voters, non response not truly random due to social desirability,
> non-attitudes, etc.) Is there likely to be some change over time?
> (history, maturation, cohort effects, vote intention formation or
> change, perception of what is likely to happen, wanting to be on winning
> side, etc.)? How reliable is the measurement process: wording,
> understanding, sequence of questions, interviewer or instrument by
> respondent interaction, etc.?

Many of these are acknowledged problems among pollsters;

Frame problems
A) Non-phone households
B) Cell phone only households
C) Reliability/Timeliness of Registration based sample

Non-response problems

But one of the biggest current concerns among pollster deal not
with these scientific concerns but with the art of determining
who will vote.

Some pollsters (Gallup for example) use as many as 7 questions to
determine whether a respondent is likely to vote while others use
two or three questions. Then they (generally) take the top
scoring X% and assume that they are likely to vote. The number
that most are using is 55% which is the proportion or registered
voters who voted in the last election.

Some pollster weight the sample to reflect what they think the
proportion of Republican to Democrats are likely to be in the
voting population other think party affiliation is too likely to
change. 

-- 
Leo G. Simonetta                                      
lsimonetta@newsguy.com

Relevant Pages

  • Re: Ohio State--ND in polls
    ... >> The point was that pollsters are indeed dumbfucks, ... > rank the sextet of ND, Ohio State, Oregon, Auburn, Miami and UCLA. ... Both are very respectable rankings, ... I think ND haters are more likely to vote realistically on ND than ND ...
    (rec.sport.football.college)
  • Re: Request for estimate: USA Protest againt political polling
    ... Some pollsters ... It would look this way if the Undecided voters ... then half of them do not vote. ... was a massive turnout in his win as governor, ...
    (sci.stat.math)
  • Re: OT: Clinton Sour Grapes. How Sweet!
    ... perhaps so samurai, but on the flip side, i think history has shown a ... certain % of voters who say to pollsters they will vote for a minority ...
    (rec.music.opera)
  • Re: Ashamed Democrats lie to pollsters :)
    ... I make a point of lying to pollsters - what I vote is for me to know and ... If enough people lie, discrediting the pollsters utterly, then maybe we can ...
    (sci.electronics.design)