Re: ID cards - What wasn't mentioned about the results of the consultation exercise

[Brian Beesley <[EMAIL PROTECTED]>]

On Monday 01 November 2004 23:20, you wrote:
> In message <[EMAIL PROTECTED]>, Roland
> Perry <[EMAIL PROTECTED]> writes
>
> >The American pollsters on TV this morning are saying that a *carefully
> >selected* representative sample of 1000 voters [asked one question]
> >will result in a +/- 3% accuracy. As one of then said "if the result
> >appears to say Bush will win by 2%, then problems with the sample could
> >mean that actually Kerry might win by 1%".
> >
> >They then said that to get 1% accuracy you'd need a sample of 3,000,
> >which is suspiciously linear, and therefore I doubt it.
>
> You are right to doubt it. Assuming the sample is a small fraction of
> the total population, then the error due to sampling effects is, to a
> good approximation, inversely proportional to the *square root* of the
> sample size. Of course, this error estimate is an underestimate, as it
> completely ignores the systematic effects due to not having a
> representative sample.

Also that the "3% accuracy" quoted is an estimate of the error - I'm not sure 
but probably an estimate of the standard deviation, in which case there's a 1 
in 20 chance that the actual estimate is out by 6%, 1 in 1000 that it's out 
by 9%. Also, with a 3% SD error in the measured support for each of the two 
candidates, the (critical) difference between Bush and Kerry is measured to a 
SD of sqrt(2) * 3% i.e. uncertain by about 4.5% at the 65% confidence level, 
9% at the 95% level.

Opinion surveys are also flawed in that few surveyors will dare to publish an 
estimate which differs wildly from previous estimates from their own or other 
similar surveys for fear for showing up statistical errors, which the 
ignorant will interpret as flaws in their methodology.

We do, however, seem to have a shift in so far as the Home Office now seems 
to agree that support for their ID scheme is less than the 80% they were 
claiming until last week. I welcome this but still doubt their methodology as 
well as their figures.

Brian Beesley