Thursday, January 13, 2011

Global Warming and Bias in Data

Here is a bit of evidence of "global warming" which got overturned when scientists discovered that like quantum physics, the observer's interaction with the observed skews the data and creates a misperception of what is truly going on:



This is one of those good news/bad news stories. It is bad news because it says the science isn't as simple as once was thought and it means that a bit of evidence of "global warming" just went "poof!". But it is good news because it means that the world is a lot more interesting than we thought and that the game of science, the struggle to understand, is more complex and more interesting than we first thought and our science really is self-correcting!

Science isn't like other disciplines where appeals to authority or the requirements of blind faith or a buying into the "secret sauce" lets an outsider become an insider and "know". Science is completely democratic. Anybody can do it. The final arbiter is fact, the cold hard facts of the real world. So in the end, science will correct for those blinded by the beauty of their hypothesis or the misdirection caused by scientific frauds perpetrated by those who want to control the outcome of science (think of the Piltdown man hoax or Cyril Burt's faked twin study data on the inheritability of IQ).

I have this soft spot for bottom-up institutions: democracy, science, evolution, the English language, etc. These are all messy processes, but they end up being the only ones that provide a solid basis for improvement and, over time, produce wonderfully complex outcomes.

I have a serious dislike of top-down institutions: orthodox religion with fixed dogma, the Académie française pontificating on "correct" French, fanatics like Hitler with their "final solutions", bureaucrats who lose the point of rules when executing a public service and who forget they are servants of the people and not petty tyrants, etc.

This is one reason why I flinch when I hear global warming types tell me the "science is settled". Science is never settled. Newton reigned for 300 years, but was overthrown. Nothing could have seemed more "settled" than Newtonian gravitation in 1900. But by 1920, a mere twenty years later, it was dust, blown apart by the new theory of Einstein.

Science is a process. Theories come and go. Nature remains. Even "the facts" become reinterpreted as the theories evolve. But nature remains. The hard fact of the way the world is remains to guide the scientific process. Tagged penguins dying at a high rate seems to indicate that there is a stress on the penguins and "global warming" is a good culprit to name. But ultimately, hard facts win out. It wasn't "global warming". It was the tags that skewed the measurement!

2 comments:

Unknown said...

RY;

Years ago.. I can't remember who it was that I heard this from, but this comedian or a friend of my dad was talking about this TV episode of Wild Kingdom where they were studying deer and logging areas; well, they came down amongst a herd of deer and tranquilized a number of them and put transmitters on them. Then they monitored the deer's movements; the deer moved to another section. This guy telling the story laughed and said, "of course they moved; you would move too if a UFO came down and shot you with a dart and left some sort of hardware attached to your body.

I always wondered about the health of an animal after being "studied" by scientists on those TV shows and documentaries.

RYviewpoint said...

Thomas: All universities in advanced countries now have ethics committees that review all proposed experiments on animals in university labs (see here). Their job is to ensure that the research is necessary, done humanely, and with no unnecessary pain for the animals. The "Wild West" days of experimenting on animals is long past. But this video shows that that isn't true in field experiments. But I'm guessing that the days of experimenters freely doing whatever "experiments" they want in the field are numbered.

This is good and bad. It is good because it ensure no unnecessary pain and that animals are treated well. It is bad because it does add a layer of bureaucracy. This undoubtedly has slowed discoveries.

The definitive experiment to prove that Helicobacter pylori caused cancers was done by researchers who used themselves as the "guinea pig" for the experiment:

The classic case of human experimentation that cut out the paperwork?

To demonstrate H. pylori caused gastritis and was not merely a bystander, Marshall drank a beaker of H. pylori culture. He became ill with nausea and vomiting several days later. An endoscopy ten days after inoculation revealed signs of gastritis and the presence of H. pylori. These results suggested H. pylori was the causative agent of gastritis. Marshall and Warren went on to demonstrate that antibiotics are effective in the treatment of many cases of gastritis.

What bothers me is that ALF (Animal Liberation Front) refuses to recognize that the scientists are now regulated and that animal experimentation is necessary for the good of humans and animals (new veterinarian drugs & surgical techniques have to be tested by experiment just like for humans so research is essential). ALF activists are fanatics and are so convinced they are "right" that they are willing to kill humans and destroy property and research as they act as judge, jury, and executioner. Sad.