The only change is the way you conducted your poll.. How about the real polls about Abortion, they have changed drastically, but that doesn’t count.. Go Figger.. :)
add in obamacare as well....
PEW Research reported last May that just about all the random sample pollsters (and even those with pre-selected polling groups) are down to a 9% response rate.
That means if you call 1000 people, you end up with just 90 of them providing you a response.
That's 1 out of 11.
A big part of the problem is that people just don't respond to calls from pollsters ~ after all, being commercial operations they have to send a signal to your phone telling you who they are ~ and you don't respond!
The problem at this point is that minority groups within the population can seize the day and take over the polling results simply by answering the phone.
Here's the way it works. Call that 1000 people your computer sampled for you. Let's say there's a group that's only 5% of the population and they always answer the pollster calls. You will get 50 of them!
. The other 95% only answer and responded at the 9% rate ~ which means they will represent only 8.55% of the total or 86 respondents.
As you can see that 5% now represents 37% of the responses, and the 95% has dropped to a here 63%~
If it's a Presidential preference poll, you'd expect something in the order of 50/50, but here'd you have 63/37!
You could be easily misled into thinking a Romacain was winning and a Gobama was losing.
Gee, that happened didn't it.
This also addresses the reason why Gallup thinks there are 3.4% gays out there ~ they asked all their subjects that question for the last year ~ and of course all the gay guys knew that! The real answer is more like .34% are gay ~ or 1/10 less BTW.