There was around a 10 point difference but the sample differences where in the 30 point range. I'm guessing / exaggerating because I don't feel like looking it up but you get the idea.
. On Wed, Jun 5, 2013 at 9:09 AM, Larry C. Lyons <[email protected]>wrote: > > To expand when doing statistical sampling you want a representative sample > of the overall population. Very large scale surveys and the census have > indicated that there are more Democratic than Republican voters. > > What you would want - equal numbers of Democrats and Republicans, is a > false equivalence. That sort of sample does not reflect the population as a > whole. Any trends or averages within such a skewed sample would not > necessarily reflect similar trends or averages within the overall > population. Hunter, Schmidt, and Jackson (1982) provide a very good > example. In a Monte Carlo simulation they pulled a series of samples and > calculated the population relationship. Those samples that were at > significant variance from the population structure tended to have the > greatest variability in their estimates of the population relationships. > > So the changes of a survey artificially boosting the number of Republicans > would tend to be less predictive than one structured like the overall > population. So seriously what would be more predictive, a cherry picked > sample or one that is structurally very similar to the overall population. > > > Hunter, J., Schmidt, F., and Jackson, G. (1982). *Meta-Analysis: Cumulating > research findings across studies.*Beverly Hills CA: Sage. > > > On Wed, Jun 5, 2013 at 9:00 AM, Larry C. Lyons <[email protected] > >wrote: > > > No Jerry that is demographics. You know how the over all population is > > made up of? > > > > > > On Tue, Jun 4, 2013 at 9:49 PM, Jerry Barnes <[email protected]> > wrote: > > > >> > >> "I remember Jerry and Sam proselytizing that the polls were biased to > the > >> Democrats. They said that the majority of people questioned were > >> Democrats, and > >> so the polls were all skewed." > >> > >> The internals for these polls showed over and over again a larger > >> percentage of Democrats than Republicans. That is bias. > >> > >> > >> "They mentioned that the 'correct' polls showed Romney winning." > >> > >> I don't know who "they" is. I didn't mention any correct polls. > >> > >> > >> I'll admit I thought Romney would win. I underestimated the willingness > >> of > >> Americans to except an real unemployment rate of 20+% and a > >> corrupt/incompetent administration. > >> > >> J > >> > >> - > >> > >> Ninety percent of politicians give the other ten percent a bad > reputation. > >> - Henry Kissinger > >> > >> Politicians are people who, when they see light at the end of the > tunnel, > >> go out and buy some more tunnel. - John Quinton > >> > >> > >> > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Order the Adobe Coldfusion Anthology now! http://www.amazon.com/Adobe-Coldfusion-Anthology/dp/1430272155/?tag=houseoffusion Archive: http://www.houseoffusion.com/groups/cf-community/message.cfm/messageid:364233 Subscription: http://www.houseoffusion.com/groups/cf-community/subscribe.cfm Unsubscribe: http://www.houseoffusion.com/groups/cf-community/unsubscribe.cfm
