8 posts Page 1 of 1
Dr. Jelinek spoke in his presentation about bias in research about the odds of a result occuring by chance, i.e. p-value. Normally, the odds seem overwhelming given the sample size, that an effect was real. But some interesting facts are being discovered about something called Regression to the Mean - As the experiment is repeated, an early statistical fluke gets cancelled out. You'd think that this means the first trial goes well, but then can't be repeated. However, for some reason it appears that it takes time for the lower results to come out (studies with initially low results would be less likely to be repeated, so we're left with the "successful" trials being repeated).

Take a look at this facinating article. Jonathan Schooler made a career-creating discovery about human memory and then saw his results decline every time he repeated the experiment which won him so much acclaim. Although it troubled him, other researchers told him not to worry, it happened all the time (not very comforting). He refused to give up and stand fast by his original research that is now heavily quoted and for which he's famous. He think's he's discoverd something more important - that somethign may be seriously wrong with the way we do research.


http://www.newyorker.com/reporting/2010 ... ntPage=all
Alex

Diagnosis: Jan 2010, OMS April 2010.
Great article Alex. I love the last line: "When the experiments are done, we still have to choose what to believe."

It's much easier to believe the science behind the approach we advocate here when you stay well....

Be well

George
That is such a spooky article. Makes you wonder what is the optimal amount of re-trialling that needs to be done in order to be assured of a reasonably accurate average. And even more worrying is can we apply that same principle to the Swank results. Are we part of the retrial?

Interesting stuff and more food for thought.
It's all so difficult and confusing. If you do a statistical analysis of MS in the world, the darker the hair colour, the less frequent the incidence of MS.
Alex

Diagnosis: Jan 2010, OMS April 2010.
We still can't get 100% definate answers or there would be little need for this website or for us to explore. I've wondered about Swank's research - he falls victim to the same issues - it wasn't properly controlled, it was subject to bias (his own desire).

But you do have to put your confidence in something. there are many cures with different data. I think it's a personal choice based on what makes sense to a person, and to a large extent, we're putting our faith in the Professor who has the disease himself and is not merely seeking fame and fortune.

The thing is, that what we're doing here is pretty much our best bet. It puts together the best advice on Medecines, Diet, Medetation, Exercise and Psychological factors. Their sum total may also be greater than the sum of their parts as well.
Alex

Diagnosis: Jan 2010, OMS April 2010.
R.E. dark hair - I think this ties in with the vitamin D thing. Fair hair and complexion is more common in populations living further from the equator, so dark hair = more likely to live closer to the equator = less likely to have MS.

R.E. regressing to the mean - yes, this is a very good point, and a classic mathematical observation that's true of absolutely any data. It doesn't mean that all successful trial results can't be repeated - just that if one's a fluke (which is less likely when it's a properly controlled trial with a large sample size, but is *never* impossible) repeat trials could show it up. Clinial trials are so expensive to organise and run, though, that repeating them is rare.
No, I meant it's hair colour:) I was just trying to demonstrate how difficult it can be to interpret data. I could be famous for discovering the hair colour MS link when everyone thought it was VitaminD.
Alex

Diagnosis: Jan 2010, OMS April 2010.
George wrote: It's much easier to believe the science behind the approach we advocate here when you stay well....

Be well

George


You'll have to allow me my doubts and Insecurities. However, amidst all my doubts, such statements and your example do help tremendously.

Thank you.
Alex

Diagnosis: Jan 2010, OMS April 2010.
8 posts Page 1 of 1

Who is online

Users browsing this forum: No registered users and 2 guests