|
Post by sandy on Nov 3, 2006 11:11:39 GMT -5
As a lot of you know next Tuesday is out national election day. For weeks now I have been listening to pollsters predict the outcome of the different races. I talk to people all across the US and agree that there is a lot of discontent with the way things are going and the blame is all falling on one person. Being associated with that person apparently is contaminating all party members. This is nothing new. It happens every election. What I don't understand is how a poll can predict the outcome of an election. They ask maybe 1000 people questions and they represent all of America? Are half of these people Republicans and Half Democrats? What about the Independents? Can this type of survey be really representative of a nation? Seems to me if you got to one area of the country you get one opinion and an exact opposite one in another area. Maybe this should be under pet peeves.
|
|
|
Post by Mark on Nov 3, 2006 22:46:38 GMT -5
What they purportedly do is take scientifically-based 'samples' of the population which can theoretically be extrapolated as being representative of the population at large. However, although these 'polls' might show trends as they are taken over time, I think their predictive value is suspect at best.
|
|
|
Post by sandy on Nov 4, 2006 10:29:56 GMT -5
In other words they are whistling in the dark.
|
|
|
Post by Mark on Nov 4, 2006 15:27:49 GMT -5
Precisely....or past the graveyard, depending on your point of view.
|
|