And here's another question: Say ya'll have convinced me that the faith IS the problem? What do we do about it? Just outlaw it? Go to war to eradicate it? Go on a conversion crusade?
All the polls are wrong, didn't you hear?
You'll have to make a convincing argument that any given poll you're using (a) is reliable and valid--from a psychometrics/measurement perspective, and (b) is externally valid..that is, some explanation of how the sample accurately generalizes to the population.So, according to you, all polls of Muslims are worthless? Really? You're going to base your argument on that?
Very weak argument that doesn't stand up to scientific polling.
Deus vult.And here's another question: Say ya'll have convinced me that the faith IS the problem? What do we do about it? Just outlaw it? Go to war to eradicate it? Go on a conversion crusade?
When you see several polls reaching similar conclusions, that builds a consensus.You'll have to make a convincing argument that any given poll you're using (a) is reliable and valid--from a psychometrics/measurement perspective, and (b) is externally valid..that is, some explanation of how the sample accurately generalizes to the population.
Of course it does. And that consensus has to be grounded, first, in reliability and validity...otherwise it means nothing.When you see several polls reaching similar conclusions, that builds a consensus.
...which means absolutely nothing. Sampling bias is an obvious source of error here. How were the polled samples selected? Without an answer to that (among other questions), the polls mean nothing. How, exactly, were the questions phrased (and, since you're reading it in English, how good is the translation)? You can make a poll give any result you want through selection bias and careful phrasing of the question (and, again, translation can make the question seem entirely different; translation rarely works with mathematical precision).When you see several polls reaching similar conclusions, that builds a consensus.
in other words...reliability and validity!...which means absolutely nothing. Sampling bias is an obvious source of error here. How were the polled samples selected? Without an answer to that (among other questions), the polls mean nothing. How, exactly, were the questions phrased (and, since you're reading it in English, how good is the translation)? You can make a poll give any result you want through selection bias and careful phrasing of the question (and, again, translation can make the question seem entirely different; translation rarely works with mathematical precision).
Dingdingding! We have a winner!Opt-in polls vs. statistically diverse population sampling.
Hillary is going to win the election!
Opt-in polls vs. statistically diverse population sampling.
Hillary is going to win the election!
Enter your email address to join: