>CurtAdams@aol.com writes:
>>Well, discussion would help, but discussion is generally with highly
selected
>>groups. You'd want people to know what other people, in the aggregate, are
>>deciding, and surveys are far better for that.
>Yes you want wide participation in the discussion, but surveys are not
>necessily better. Consider the wide use of "focus groups" in marketing.
Discussion and focus groups are often necessary if the topic is something
people don't normally think about. But aren't focus groups often selected by
the researcher, either to reflect a sub-population, or to gather a more
represtentative sample, rather than self-selected?
>>Another big problem with forecasting or discussion is that people often
>>change their minds about what they want. It's difficult for a 20-year-old
in
>>college who's never worked anything more than part-time casual work and
who's
>>never had a significant relationship or children (an awful lot of
>>20-year-olds these days) to really have any good idea what they will want
at
>>40.
>Yes, but our predictions about aggregates may be better than our
>predictions about individuals.
True. I was thinking more along the lines of how anybody could know what
modern 20-year-olds (either individually or in the aggregate) will be like at
40. The 20-year-old won't know; they haven't been there, and as I outlined
above, their predictions are suspect. The 40-year-old won't know; they and
their peers were different at 20 (insofar as they're the same, we have
neither a change nor a problem).