- Thread Author
- #1
Michchamp
Well-known member
- Joined
- Aug 4, 2011
- Messages
- 34,212
If you don't pick and choose the people you poll beforehand so that the poll results show that Romney is leading, your poll demonstrates "liberal bias." link.
For example:
picking 1,000 people out of the phone book at random, polling them, and noting that 500 identified as democrats, 400 republicans, and 100 as independents, and the poll shows Obama up 55%... well that's not good, because we don't want to live in a world where more people identify as democrats than republicans, even if true, so that would be:
LIBERAL BIAS.
pre-selecting 1,000 people that make up 50% republican ID'd voters and 50% democrats, and finding that the polls show the candidates in a dead heat?
BETTER, BUT NOT GREAT
picking 1,000 people out from a conservative, rural area, who are NRA members and finding Romney is up 95% to 5%?
YEEEHA!!!! NOW THAT THERE'S A POLL, VERN.
For example:
picking 1,000 people out of the phone book at random, polling them, and noting that 500 identified as democrats, 400 republicans, and 100 as independents, and the poll shows Obama up 55%... well that's not good, because we don't want to live in a world where more people identify as democrats than republicans, even if true, so that would be:
LIBERAL BIAS.
pre-selecting 1,000 people that make up 50% republican ID'd voters and 50% democrats, and finding that the polls show the candidates in a dead heat?
BETTER, BUT NOT GREAT
picking 1,000 people out from a conservative, rural area, who are NRA members and finding Romney is up 95% to 5%?
YEEEHA!!!! NOW THAT THERE'S A POLL, VERN.
Last edited by a moderator: