> If I recall what I have heard/read about polling techniques correctly,
> the accuracy of the polls wouldn't suffer from an increased refusal rate
> provided that they still got a large enough sample (not hard to do since
> you only need a few hundred or thousand for a national poll) and could
> adjust the results by assigning the correct weights to the various
> categories of respondents.
>
> But the polling organizations still seem to worry about running into
> more refusers. Maybe it biases the results in some way that can't be
> corrected for, or just because it makes the public *think* the polls are
> less accurate.
>
> But I think that the more important problems, as ever, are that the way
> the questions are worded greatly affects what answers one gets, that
> most reported polls hardly ever get very deeply into *why* the
> respondents answer as they do, which is usually the most interesting
> aspect of a poll, and that the way the polls are reported in the news
> media evaporates most of what is left of interest in them.
I wonder if the higher refusal rate increases the cost of the polls, and if that isn't problematic for the pollsters.
-joan
> Jon Johanning // jjohanning at igc.org
> ________________________________
> How good bad music and bad reasons sound when we march against an enemy.
> -- Nietzsche
>
> ___________________________________
> http://mailman.lbo-talk.org/mailman/listinfo/lbo-talk
>