Almalieque wrote:
someproteinguy wrote:
Almalieque wrote:
someproteinguy wrote:
Almalieque wrote:
Are you telling me that you can accurately determine a population's opinion on a highly divided subject by only polling a fraction of 1% of the total population?
Any statistician can give you an answer for that one. Then while you're at it ask a politician how to take the results out of context.
Good thing there's a difference between mathematicians and statisticians. Stats are done to social standards. There is no bias in math.
Um what?
As much as I don't want to say it, you're going to have to elaborate on that. Last time I did a T-test I don't remember there being an "unless you are gay" condition...
Being a statistician is like being a Physicist or Chemist. Although they use high level math, their focus is on the APPLICATION of the math, not the math itself. If you've taken higher level math courses with these sciences at the same time, you'll realize that those scientists develop procedures so that everyone is on the same page. To the mathematician, it's irrelevant.
A good example is the whole "heat exits the room, not cold air coming in". That's important to a Physicist when doing calculations in order to be consistent with the negative sign. For the mathematician, it doesn't matter where the negative sign goes.
It is not feasible for a statistician to poll the majority of the population, this is why there are multiple types of polls (with pros and cons) and there is "acceptable risk" (for a lack of a better term) that all statisticians use in order to be on the same page. However, that does not change the actual numbers in reference.
Well ok, so where's the bias?
Are people not polling a group that is representative of the larger population?
Are the scientists not choosing the right test?
Something else?
There's got to be something here other than labeling heat as positive, or republican as red...