Let's do our own poll

[groverblue]

I say let's do our own poll. I'd be willing to coordinate our efforts. In the meantime, here is a poll on Twitter:

https://twitter.com/johnmanko/status/670768182613958656

[CPUd]

I think there was some talk of doing this in 2012, someone knew a company who would do it for around $2K, but I forget the thread.

[fatjohn]

If it gets done, i am for mobile only polling.

[jkob]

It would be very interesting to commission our own polling, I wish it was done in 2012 before Iowa because maybe it would showed something different than a "Santorum surge"

[Crashland]

It's easier to just look at fundraising and unique donors, which pretty much reflects exactly what the national polls are saying.

[Peace&Freedom]

It would be very interesting to commission our own polling, I wish it was done in 2012 before Iowa because maybe it would showed something different than a "Santorum surge"

Polls by prestige survey firms cost thousands to do, based on the quote I got my trying to start one through Zogby in mid-2007. The Santorum surge was based on an extrapolation from the last two days of the final Des Moines Iowa Register poll, and so the media designed to make it sound authoritative to hype it up. If another poll had been done at the time showing something different, the MSM would have ignored it, since the Des Moines poll already gave them cover to hype up Santorum while ignoring Ron.

[groverblue]

If it gets done, i am for mobile only polling.

Can't do mobile polling. It's against FCC regs. I think maybe only candidates can do mobile polling.

[groverblue]

It's easier to just look at fundraising and unique donors, which pretty much reflects exactly what the national polls are saying.

hmm, I should be able to do something using public data next weekend. Great idea.

[Foreigner]

The current polls are accurate in terms of their methodology.

If one commissions a private poll you would get the same results unless you change the methodology. The methodology must still be representative of the likely electorate, which in the case of Iowa isn't a very large portion of the population at large.

It's easy to say: poll more young people on their cell phones; but would they pick up and answer?

If one has to call 4000 young people to reach 200 of them, then it gets expensive to conduct such a poll. And, is the 200 who are inclined to pick up the phone out of the 4000 really representative? And who many of them will likely participate in the caucus? Maybe you need to call reach even more young people to get the sample group you really want.

In general, a lot of studies have a bias at the start, since participation is voluntary. Are people who run to participate in all kinds of studies really representative of the majority who think it would be a vast of time for them to do so? Still, studies must be conducted, but it's something to keep in mind.

[Mr.NoSmile]

Hardly a reputable poll with just those options and none of the other candidates on the right. It'd be like if Trump released a poll with the only options being 'Yes' and 'Maybe.'

[RandFTW]

someone should conduct a poll asking whether people would rather elect a president that will grant them immediate gratification without regard for future benefit or a president that will focus on long term benefits of the country. I don't think most people realize this is the true, simplistic difference between Rand Paul and the others.