How To Import Raw Political Data For Crunching
Ed Pegg writes "For those that want to get their fingers stained red and blue with actual political data, resources beyond 538 and pollster can be accessed. In a blog item for Wolfram Research, Jeff Hamrick gives step by step details for how to import raw data from Mason-Dixon, Rasmussen, and Quinnipiac. Then he uses Mathematica to analyze the political data." Related: Slashdot developer Pudge presented at OSCON in July his own approach to gathering Washington-state polling data for analysis [PDF].
...to import bullshit?
How about importing raw political candidates for crunching instead? If sacred cows make the best hamburger, what would we make with politicians, chum?
Kwisatz Haderach
Sell the spice to CHOAM
This Mahdi took Shaddam's Throne
Pudge presented at OSCON in July his own approach
Considering all of the hardcore conservative journal entries that Pudge has made in just the past few weeks, I'm not sure I want to know how he comes up with his conclusions.
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
For those people that don't want to install mathematica you could go to fivethirtyeight.com and get updated polls. I got this from the typewriter xkcd.
This doesn't seem like a serious forecasting project, just a bit of topical advertising fluff thrown together by a couple of Wolfram staffers. It might be fun to download and play with if you have Mathematica already, but it doesn't make me want to rush out and blow two and a half grand on propriety software.
[ ]Half Empty [ ]Half Full [x]Twice as big as it needs to be
If you want to import data from the *money* in politics, which I generally find more amusing/entertaining check out the Federal Election Commission's FTP site
By a suspiciously large margin.
If you want to see some poll meta-analysis that is pretty rigorous check out election.princeton.edu.
1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597 2584 4181 6765
What I'd really like to see is if the data were replaced with purely random data (or as close as we can get) would the people "analyzing" it get any different results, really? Or would they keep finding the trends they are looking for, do you think?
Nothing to say here... move along
You could make something more useful than chum, pal...
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
As a student of mathematics, I say that it is much more difficult to manipulate data to legitimately suggest different results that most would suspect. As they say, the numbers never lie. Of course, the statisticians might lie. Or the statistics might lie, in the sense that there is always a chance of a biased study. But the numbers never imply causation, merely correlation.