[EMAIL PROTECTED] wrote:
A "margin of error" is often cited in political and other public opinion
polls; this is a very loose statistical measure, but one that seems to
have stood the test of time in such matters.
It's the reciprocal of the square root of the sample size. In this case,
the number of respondents was 99, so the margin of error is 1/sqrt(99),
which is a hair over 10%. On this basis, there's no statistical
significance between 49% and 44%.
this is a good explanation, but also bear in mind that in
this case we cannot even tell (nor can the dp) if the number
of respondents was 99 different individuals, or one
individual (or office, or organization) with access to 99
different computers, or something in between.
please, everyone, remember that these anonymous online
'polls' merely count computers, not people!
- - -
I believe the term for this kind of poll is "voodoo poll",
but others here who have more expertise might be able to
explain it better.
http://www.doubletongued.org/index.php/citations/voodoo_poll_1/
http://ukpollingreport.co.uk/blog/archives/1117
- - -
meanwhile, have fun:
Do you think Obama should be the next
president of the United States?
___ yes
___ no
___ maybe
..................
UNIVERSITY*CITOYEN
----
You are receiving this because you are subscribed to the
list named "UnivCity." To unsubscribe or for archive information, see
<http://www.purple.com/list.html>.