Wednesday, June 30, 2010

Daily Kos Poll Data Faked

I have used two stories on this site that came from polls done by the Daily Kos. Now it turns out that the company that the Daily Kos used for its polling was faking the data. TechDirt has this to say:
You may have heard by now that the political website Daily Kos has come out and explained that it believes the polling firm it has used for a while, Research 2000, was faking its data. While it's nice to see a publication come right out and bluntly admit that it had relied on data that it now believes was not legit, what's fascinating if you're a stats geek is how a team of stats geeks figured out there were problems with the data. As any good stats nerd knows, the concept of "randomness" isn't quite as random as some people think, which is why faking randomness almost always leads to tell-tale signs that the data was faked or manipulated. For example, one very, very common test is to use Benford's Law to look at the first digit of data in a data set, because in a truly random set, the distribution is not what people usually expect.
The two stories I used were on this post and this one. Sorry if I inadvertently misled anyone. Interestingly, several researchers, including the great statistician R.A. Fisher, have suggested strongly that the father of inheritance studies, Gregor Mendel, cooked his data. Piegorsch1 writes this:
Part of Fisher's effort in the reconciliation of Mendelism and (biometric) Darwinism was the recognition that only superficial scientific scrutiny had been directed at Mendel's article by both its proponents and opponents (Box, 1978, p. 297). By the early 1930s, Fisher had undertaken to critically review Mendel's original article, going further-in a sense-than those earlier researchers. His conclusions included an astonishing calculation: Mendel's data appeared to fit his own expectations and proposed theories far better than expected by chance alone. The quality of the data (but not the theory itself), and even Mendel's historical reputation were soon called into question by the scientific community. Ironically, after reconciling one scientific controversy on the effects of heredity in evolution, Fisher's research helped spark another!
This is a controversy that continues to this day. It has been mitigated somewhat by the fact that most of Mendel's theoretical constructs with regard to inheritance turned out to be true.

1Piegorsch, W. (1990). "Fisher's contributions to genetics and heredity, with special emphasis on the Gregor Mendel controversy." Biometrics 46(4): 915-924.

Now playing: Weather Report - Cucumber Slumber
via FoxyTunes

No comments:

Post a Comment