Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
At the end of the day, 2008 was historic for turnout. So, the question remains. Is it a mistake for these firms to use 2008 to help determine sampling size and breakdown?
No.
The estimate is adjusted up or down depending on Americans' reported interest in the election and voting intention.
I'm not sure where your getting this from but as an example Quinnipiac does a straight survey with no weighting and Bloomberg does a weighting that matches the census. Neither uses previous elections. I can go on and on but I don't really want to be long winded.
Okay, then it would seem some polls versus others are better indicators of reality or no? I was just thinking about how much reliance the media, etc puts on polls and how, in the past, they have sometimes been so terribly wrong. Thus, that led to my questioning why this has occurred since I have an interest in the computational models. This election, the polls have been quoted again and again by both sides. Thus, it should make us all naturally curious as to the validity of the polling methods.
You don't know Ohio. The only part of the state which is growing is the Columbus area, which is increasingly liberal by the year due to relocatees and the overall intellectual bent of the population. Northeastern Ohio (Cleveland, Akron, Youngstown) has always been blue-collar Democrat. Couple it with the cities (not metros) of Cincinnati, Dayton and Toledo, along with sporadic support throughout the state, you have a margin of victory at least that of 2008. The only Republican strongholds in the state are southwestern Ohio (outside of the cities proper) and the rural areas, and they're not enough to push NoBama through.
You do realize there has been a foreclosure crisis in Ohio? You do realize registrations are down in the state about half from in Cleveland and surrounding counties?
Your guys drove the bus into the ditch and now you expect us to put them behind the wheel again. With the same roadmap that got us in the ditch in the first place...tax cuts, wars, deregulation.
The estimate is adjusted up or down depending on Americans' reported interest in the election and voting intention.
But again, there is an assumption made in the computational model as a result. Very interesting (okay maybe not for the average person). I always enjoy breaking down any kind of 'statistics say' type of claims to see how they actually came up with that. On a side note, I wish more Americans had an interest in mathematics but that is off topic.
You need to educate yourself on the computational models used by the major polling firms.
I think you are the one in need of education. But hey, do you want to discuss sampling, or not? Because you've been doing a fine job of dancing around the question.
At the end of the day, 2008 was historic for turnout. So, the question remains. Is it a mistake for these firms to use 2008 to help determine sampling size and breakdown?
These liberals are still scratching their heads. Not to mention the fact that they increase or decrease the expected turn out for each party based on that parties turn out in the previous election. That's where the Obungler is getting these 9 point advantages in some polls.
Take a basic college statistics class libs, any teacher worth his/her salt will have a lesson on polling.
It was actually in listening to the pollsters themselves as they've been interviewed by various news outlets. I would think a better determination of sampling size would be to look at an average over all elections instead. Perhaps it's not considered 'big' news or something that has been all that newsworthy but as a self-professed math geek, I found that information quite interesting. Thus, you hear the pollsters themselves giving a bit of caution about exact numbers. I suspect they fear if the election doesn't follow their polls, they have an out. After all, as a polling group, your validity lives and dies with how closely your polls matches actual results. Oh and of course, we're not even getting into the adults vs registered voters vs likely voters issue. Just something interesting to note is all.
Once again, why don't you answer the question?
Who is basing their sampling on previous elections?
At the end of the day, 2008 was historic for turnout. So, the question remains. Is it a mistake for these firms to use 2008 to help determine sampling size and breakdown?
Yeah, Gallup's voter model predicts greater than 50% voter turnout.
Care to explain how they are basing their sampling on previous elections. That's what you said earlier. So, tell us who is doing that.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.