Question for an anthropologist on evolution and statistical testing?

Question

I have a question on Peason/Neyman/Fisherian statistics and evolution. If I were testing for the existence of evolution using regression, such as a time series rate of change, then the most rational null hypothesis is the "no effect," hypothesis, that is, mu=0. In the category where mu=0 is intelligent design, creationism, but also other weird but possible naturalistic explanations. If the null is rejected then by modus tollens, to some degree of confidence, intelligent design is rejected by the data without a need to "assume" naturalism or other materialistic assumptions. Further, as Frequentist methods guarantee an alpha level of coverage, they are a worst case test statistic (assuming a proper statistic was used of course). This would be the distribution that most favored the null, to guarantee coverage. So using the data, with basic assumptions like Kolmogorov's axioms or probability, one would arrive at the rejection of creationism without resorting to naturalistic assumptions. What is the flaw in this argument? Also, sorry to use your professional time on this, but I cannot walk through the flaw. One could extend this, with a Bayesian decision theory framework to modus ponens as well, but only if you include additional assumptions. Criticism very much desired.

Update:

Note I am only interested in the strength of the argument on its own merits, I am not concerned with whether or not it is received by the audience as valid. I doubt creationists would "get it," as there would be a strong emotional loss from accepting it. Also, sorry about the typo on Pearson's name.

Batlow · Accepted Answer

It won't work.  You're trying to use very quantitative methods to analyse very nebulous, qualitative data.

It's a bit like asking "Can we use Fourier Transforms to decide whether Obama is a good President?" or "I want to use a Hodges-Lehmann-Sen estimator to see if we should have gun control".

Inferential statistics can be an extremely powerful tool for analysing specific sets of numerical data, which could be used to support a more general argument (indeed, the American NRA lobbied to stop medical researchers doing statistical research on gun deaths, precisely because of the evidence it was bringing up). But statistics alone won't settle disputes like evolution vs intelligent design.

The branch of science called Bioinformatics applies the full range of statistical analysis - inferential and others - to biological data. A large portion of this field deals with genetic and evolutionary material. If there was a way to resolve the intractable "evolution vs creationism" debate using a statistical approach, one of the many thousands of bioinformatics practitioners would have already done it.

Andrey Kolmogorov's insights into turbulence and stochastic processes have very wide applicability; I agree we should use them more in trying to understand the world around us. But they can't solve this problem, not directly.

Hope it helps.

Anonymous · Answer

The gigantic time factor involved in the evolution of species, even if one limits the consideration to only consecutive generational individual reproductive events; weakens the argument by making the frequency rate / aspect so small that the statistical analysis may mathematically fail.

Anonymous · Answer

I am really impressed that BATLO gave a coherent answer to an incoherent Q

Math is internally consistent ( we hope) but does NOT have to apply to physical reality

statistics is the science of SWAG ( scientific wild *** guessing)

Trending News

Question for an anthropologist on evolution and statistical testing?

3 Answers