Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

If the experiment reliably produces the effect that they are investigating, then they have at least a million dollars in their future.

http://web.randi.org/the-million-dollar-challenge.html

But the sample size planned for this study is tiny, and the study won't really have the statistical power to demonstrate a true effect, even if the groups differ in outcomes (which I rather doubt they will).

AFTER EDIT: I see a downvote, presumably indicating disagreement, has come in. It would be a courtesy to onlookers here (and to me, for that matter) to explain what part of my comment you disagree with, if you please. To be clear, I genuinely think, based on much reading of medical research, that the effect the investigators are looking for will not be found as a replicable effect, even if their preliminary low-n study shows a difference in outcomes among the different study groups. Extraordinary claims require extraordinary evidence, and to claim that tricking people into thinking they live at an earlier time will reverse the disease progression of cancer is to make an extraordinary claim (which is why I think this claim may be within the scope of the James Randi paranormal challenge).

P.S. Has any onlooker calculated how large a difference in outcomes would have to be observed in a study with only 24 participants in each group to convincingly show that the treatment effect is genuine? Statistical power is an issue that is often neglected in published preliminary research.



Is this the sort of thing that's in the scope of JREF's challenge? It says "psychic, supernatural, or paranormal ability". If this experiment does show a result from manipulating mental age with a throwback environment, we wouldn't assume that result derived from a paranormal cause. It would be a new psychosomatic effect where the mind can affect the body, but it would be through normal biological and chemical means, not the supernatural evidence JREF is looking for.


Yeah, it sounds like this is talking about (effectively) a variant on the placebo effect.


I don't think the JREF would accept them as an applicant. The JREF specializes on much smaller scale tests where they can stop shenanigans and allow for very low p-values. For example, they'll task a dowser with determining which of ten boxes filled with sand contains a gold nugget and repeat that six or more times so they can hit a p-value of 0.000001 or lower. Requiring a medical study to achieve that kind of p-value would cost more than the prize, because there's so many more variables in play.

Also, they'll probably simply disagree that totally changing someone's environment counts as a paranormal way to influence cancer outcomes.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: