[EMAIL PROTECTED] wrote:
>Robert J. MacG. Dawson <[EMAIL PROTECTED]> wrote:
>
>: If I attempt to survey 1000 people and 950 answer, of whom 600 give a
>: positive response, I can consider the extremes of 650 in 1000 and 600
>: in 1000, create confidence intervals, and say (eg) that _in_any_case_
>
>Don't follow this. Why 650 in 1000, maybe 950 in 1000?
The reason why "650" is that all but 50 of the the 1000 already responded. Of
the 950 who responded, 600 were positive. If all of the remaining 50 were
positive, the number positive would be 650 (it can't be any higher); if all 50
were negative the number of positive would stay at 600.
>
>: the proportion of positives in the population at large is over 50%. But
>: if I attempt to survey 1000 people and 100 answer with 60 positives,
>: I can only consider bounds of 60 in 1000 and 960 in 1000. Neither of
>
>Also have trouble here. 940 in 1000 instead? Thanks, Adam
Similar reasoning to above. If all of the 900 who didn't respond were
positive, the total positive would be 900+60=960; if all 900 were negative, the
number positive would stay at 60.
These possible upper and lower bounds serve as a sort of sensitivity analysis.
As you can see, if almost all those surveyed responded you are much more
certain about the proportion of positives than you are if you have a high
percentage of non-responders.
Hope this helps,
Dan Nordlund
>
>: these extremes is impossible if the response and the probability of
>: responding are strongly correlated. With such numbers I can do nothing;
>: the correct outcome is a failure to reject any null hypothesis and a
>: conclusion that the data do not support any definite conclusion at all.
>
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
