Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Just 11% of 53 cancer research papers were reproducible

nature.com

Post reply on HN
Print view View original

11–20 of 184 posts

Re: Just 11% of 53 cancer research papers were reproducible

#11

Post by JoachimSchipper »

viggity wrote: post #7

After hearing similar things about psychology papers, this is rather disconcerting. This is why I am a climate change skeptic, I don't know whether it is happening due to CO2 or not, but I am confident that science can't be confident when they can't do control experiments. You can't control for any variable when it comes to the climate, let alone all the reasonable ones. We have infinitely more capabilities to contro…

You took an almost entirely unrelated story and made it about your political hobbyhorse. Please keep politics off HN.

Re: Just 11% of 53 cancer research papers were reproducible

#12

Post by codeulike »

Most published scientific research papers are wrong, according to a new analysis.

http://www.newscientist.com/article/dn7915-most-scientific-p...

Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124. doi:10.1371/journal.pmed.0020124

http://dx.doi.org/10.1371/journal.pmed.0020124

Re: Just 11% of 53 cancer research papers were reproducible

#13

Post by campnic »

viggity wrote: post #7

After hearing similar things about psychology papers, this is rather disconcerting. This is why I am a climate change skeptic, I don't know whether it is happening due to CO2 or not, but I am confident that science can't be confident when they can't do control experiments. You can't control for any variable when it comes to the climate, let alone all the reasonable ones. We have infinitely more capabilities to contro…

I realized that by replying i'm probably encouraging you to continue posting this non-sense. I agree with Joachim below, this is barely related to the article.

Re: Just 11% of 53 cancer research papers were reproducible

#15

Post by gordaco »

rajeevk wrote: post #6
wheaties wrote: post #4

This is one of the reasons I went into math and software. In math there is only black and white, provably true and patently wrong. Still, part of the reason this is so is that to show any results everything must be laid on the table. CS and Math papers are therefore open by design and the world has benefited. I hope the rest of science will move in this direction.

CS has similar problem, specially in AI fields. Researches often claim the accuracy rate more than 95%, but in reality that is much lesser (in fact less than 10% in most of the claims).

Having worked in computer vision and robotics, I can attest to this.

Re: Just 11% of 53 cancer research papers were reproducible

#16

Post by jamessb »

wheaties wrote: post #4

This is one of the reasons I went into math and software. In math there is only black and white, provably true and patently wrong. Still, part of the reason this is so is that to show any results everything must be laid on the table. CS and Math papers are therefore open by design and the world has benefited. I hope the rest of science will move in this direction.

There are 'proofs' that contain errors, just as there are science papers that reach incorrect conclusions.

Also, it's not necessarily easier to verify a proof than it is to repeat an experiment (or perform additional supporting experiments), especially for complex proofs which very few people know enough to completely understand (eg. of Fermat's last theorem, or the Poincare conjecture).

Re: Just 11% of 53 cancer research papers were reproducible

#17

Post by 5kg »

rajeevk wrote: post #6
wheaties wrote: post #4

This is one of the reasons I went into math and software. In math there is only black and white, provably true and patently wrong. Still, part of the reason this is so is that to show any results everything must be laid on the table. CS and Math papers are therefore open by design and the world has benefited. I hope the rest of science will move in this direction.

CS has similar problem, specially in AI fields. Researches often claim the accuracy rate more than 95%, but in reality that is much lesser (in fact less than 10% in most of the claims).

Yes, that's why I no longer consider pursuing PhD in CS. It's a waste of life to reproduce non-reproducible papers that is full of buzzword.

Hopefully, more journals and confs will follow the Open Access and Reproducible Research model like IPOL[1].

> Each article contains a text describing an algorithm andsource code, with an online demonstration facility and an archive of online experiments. The text and source code are peer-reviewed and the demonstration is controlled.

[1]http://www.ipol.im/

Re: Just 11% of 53 cancer research papers were reproducible

#18

Post by kailuowang »

viggity wrote: post #7

After hearing similar things about psychology papers, this is rather disconcerting. This is why I am a climate change skeptic, I don't know whether it is happening due to CO2 or not, but I am confident that science can't be confident when they can't do control experiments. You can't control for any variable when it comes to the climate, let alone all the reasonable ones. We have infinitely more capabilities to contro…

The possibility of CO2 causing global warming isn't 100% but it isn't very low either. It is the risk we are talking about, the risk of doing nothing and let the man made green house (if there is one) turn earth into an irreversible disastrous environment. Most CO2 emitting energy sources are not sustainable anyway and many of them emit other proved pollution as well. There is really nothing lost to going green energy, other than the cost of having some CO2 emitting energy reserved there.

Re: Just 11% of 53 cancer research papers were reproducible

#19

Post by kailuowang »

codeulike wrote: post #12

Most published scientific research papers are wrong, according to a new analysis. http://www.newscientist.com/article/dn7915-most-scientific-p... Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124. doi:10.1371/journal.pmed.0020124 http://dx.doi.org/10.1371/journal.pmed.0020124

Is this analysis a published scientific research?

Re: Just 11% of 53 cancer research papers were reproducible

#20

Post by pfortuny »

wheaties wrote: post #4

This is one of the reasons I went into math and software. In math there is only black and white, provably true and patently wrong. Still, part of the reason this is so is that to show any results everything must be laid on the table. CS and Math papers are therefore open by design and the world has benefited. I hope the rest of science will move in this direction.

Not to be too picky (am a mathematician too) but you ought to read about the history of the theory of "limit cycles" of planar vector fields and Dulac's "Theorem". Quite a lesson.

Edit: sorry, could not help writing this later (was on my iPhone before). The thing is that Dulac 'solved' an important part of one of Hilbert's problems. Was a 'proof' for about 60 years.

Post reply on HN