Science

"A Scientific Look at Bad Science"

"What recent research says about fraud, errors, and other dismaying academic problems."

Related: "Science Isn’t Broken: It’s just a hell of a lot harder than we give it credit for". Includes a very clever simulation exercise, "Hack Your Way to Scientific Glory".

Also related: "Many Psychology Findings Not as Strong as Claimed, Study Says" and "A Deep Dive Into the Blockbuster Study That Called Into Doubt a Lot of Psych Research". (I'll bet you the best drink in the house that these problems currently affect psychology and sociology more than economics.)

And still more, on a related but separate problem: "64 more papers retracted for fake reviews, this time from Springer journals".


"Easy DNA Editing Will Remake the World. Buckle Up."

Wired article that tries, with some success, to explain the CRISPR technique you may well have seen discussed recently.

Start your dreams. And maybe your nightmares.

Related: Steven Pinker argues that bioethicists should "[g]et out of the way" in "The moral imperative for bioethics":

In the other direction, treatments that were decried in their time as paving the road to hell, including vaccination, transfusions, anesthesia, artificial insemination, organ transplants, and in-vitro fertilization, have become unexceptional boons to human well-being.

Biomedical advances will always be incremental and hard-won, and foreseeable harms can be dealt with as they arise. The human body is staggeringly complex, vulnerable to entropy, shaped by evolution for youthful vigor at the expense of longevity, and governed by intricate feedback loops which ensure that any intervention will be compensated for by other parts of the system. Biomedical research will always be closer to Sisyphus than a runaway train — and the last thing we need is a lobby of so-called ethicists helping to push the rock down the hill.


Words to remember

From a review of Nina Teicholz's The Big Fat Surprise: Why Butter, Meat and Cheese Belong in a Healthy Diet by Donald J. McNamara in The American Journal of Clinical Nutrition, July 2015:

This book should be read by every nutritional science professional as a guide to risks of hubris and the unquestioning belief in whatever the conventional wisdom of the day is and to the consequences of basing public policy on belief as opposed to evidence of positive, beneficial effects. All scientists should read it as an example of how limited science can become federal policy, which may, in in the long run, be harmful when the basic tenets of science, skepticism, and consistent questioning are set aside to appease the powerful voices convinced that we must do something (even if we do not have the proof that that something is the right something).

(Sorry, no link because the piece is behind a paywall.)


"The Perfect Scientific Crime?"

The author makes a good point. (Read the article for the supporting case.)

An almost undetectable fraud would be to conduct a real experiment, and involve other people in it, but to control data management yourself, and substitute convincing fake or edited data for the real measurements. 

Related to science's current problems: "Top 10 ways to save science from its statistical self".


"Book Review: Tom Stossel's 'Pharmaphobia'"

"Yet, despite the book's extraordinary well-documented and indisputable examples of real progress, the negative impact of the "anti-innovation"  industry—which has used every trick in the book to slow progress—becomes just as clear. By far, the most potent weapon in the arsenal of the conflict-of interest-movement is the fabrication of the myth that the process that leads to innovation is inherently dishonest and corrupt."