Recalling yesterday’s post on some fraudulent science in Alzheimer’s research, there is one more comment need be made. Punching through to the original story covered in the post linked, there is an interesting notation:

But the Science investigation revealed evidence that the Nature paper and numerous others co-authored by Lesné, some listing Ashe as senior author, appeared to use manipulated data.

Now bear in mind the paper being retracted is 20 years old. And the key word in that pull quote is “manipulated.” Though the article does not say it specifically, the term “manipulated” Shows the problem with the statistical nature of so much of research these days that I have mentioned in these spaces many times.

More often than not, when you read these papers, you see the results of statistical analysis of data, but not the underlying data because the data is simply too voluminous to reproduce, it just resides on computers somewhere. Lots of studies these days are called “meta” studies where they analyze the analysis of a number of studies or combine the data from a number of studies. This makes it very easy to “manipulate” the data since the data proper is never reviewed. Which means it is easy to produce a desired result rather than look at the result you get, which is often inconclusive. Meta studies then compound the manipulation many times over.

The particular study in this story is not that statistically complex. But some things are. Climate change studies in particular are entirely statistical analysis of data from many sources of various levels of quality.

I don’t want to get too deep into the technical weeds here. The point is simple – this study stood for twenty years because it was easy to manipulate data and to cover up the manipulation. The next time someone tells you, “The science is settled,” don’t bet on it.

The post On More Word on Science appeared first on The Hugh Hewitt Show.