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

You Should Downvote Contrarian Anecdotes

thobbs.github.com

Post reply on HN
Print view View original

141–150 of 158 posts

Re: You Should Downvote Contrarian Anecdotes

#141

Post by beagle3 »

stephencanon wrote: post #61
ars wrote: post #47

Earlier quoted context omitted.

> Something is rotten in the core of science. It's not science that is the problem. It's that biology considers a 95% confidence sufficient. Considering how many studies are done each year, this virtually guarantees incorrect results. The reason they do that is that it's impossible to get better results, they just can not do enough trials. So they are stuck.

95% confidence would be fine (or very nearly fine) if a requirement for funding/publication would be that all data, including negative results, be published as well.

That would only be true if

a) all data, everywhere in the world, including negative results, was published regardless of funding/publication.

b) someone actually looked at that data, normalized it, and used it to assess the real significance of every result, in a sane manner (e.g. by using a bayesian inference with some reasonably behaving universal prior).

Neither a, b will ever happen, and both are essential.

(note: publication of all data is not a sufficient requirement: if 20 independent labs each do the same random experiment, one of them is expected to have a 95% confidence, and when they publish all their data, it consists of that one experiment that seems legit. This _will_ and _already does_ happen by chance)

Re: You Should Downvote Contrarian Anecdotes

#142

Post by phren0logy »

tomjen3 wrote: post #92
phren0logy wrote: post #49

Earlier quoted context omitted.

>Also, considering that medicine is at the stage of alchemy and that doctors simply have no idea what long-term effects these vaccines have on our immune system, some questions do have to be asked. What the hell are you talking about? There is probably no single more life-saving intervention in medicine than vaccines. It is true that a small number of people have a bad reaction to them, but more people have a bad rea…

If some people have bad, even fatal, reactions to vaccines shouldn't the physicians first duty (do no harm) mean that they should separate the people who would suffer sideffects from those who wouldn't?

In the US, we do a screening for risk factors for reactions, and review information needed to give informed consent.

Re: You Should Downvote Contrarian Anecdotes

#143

Post by nessus42 »

As someone who has had his ability to work saved due to anecdotal evidence, I could not disagree more with the OP. Well, sure, if someone has anecdotal evidence that the theory of gravity is wrong, or that 1 + 1 = 4, by all means downvote them, but not all scientific evidence is on such solid ground.

E.g., I've been prescribed medication and ended up with a side effect not listed on the drug literature. The doctor told me that the side-effect must be psychosomatic. Later scientific evidence revealed that about 50% of patients given the drug experienced the same side effect, but that it had been previously underreported. Who's to say that researchers would have even bothered to research the side effects more thoroughly if they hadn't paid attention to the fact that the anecdotal evidence contradicted the scientific evidence.

Or remember when the scientific evidence seemed to indicate that a high carbohydrate, low fat diet was the healthiest choice? Should I have ignored my anecdotal evidence that that diet made me feel like crap?

Science sometimes gets itself into harmful orthodoxies. See Thomas Kuhn for more info on this if you are unfamiliar. One example of this is Behaviorism in Psychology. In this field, it was scientific orthodoxy for many decades that Behaviorism was scientific and Cognitive Psychology was not because Behaviorism was based on only quantifiable, measurable data. It took Chomsky to point out the idiocy of this orthodoxy, thereby breaking the orthodoxy, allowing science to progress.

Re the anecdotal evidence that saved my career, I've read here that there is no scientific evidence that ergonomic keyboards can help prevent or ameleroate RSI. I am 100% sure, however, that the Kinesis Contour keyboard saved my ability to type. All I have to do to know this for sure is listen to my body. Furthermore, I personally know about a dozen programmers who feel similiarly. I'm sure that someone will pipe in that this is almost certainly the placebo effect. If that's the case, then Kinesis makes the world's very best placebo, as placebo-like things have done precious little for me in any other area. In any case, even if it were the placebo effect, what does it hurt anyone to ignore the putative scientific evidence and try out a Kinesis keyboard for themselves to see if it provides them with relief?

The idea that posts should be downvoted for recommending such ergonomic keyboards is insane.

Re: You Should Downvote Contrarian Anecdotes

#144

Post by vannevar »

Symmetry wrote: post #124
vannevar wrote: post #91

You shouldn't downvote to censor relevant information, period. You should only downvote if the material violates site guidelines. Downvoting to disagree is lazy, substituting easy censorship for thoughtful response. Instead, post your counter-argument or upvote an existing opinion that already expresses your position.

Censoring relevant information is, of course, never good but the authors contention was that the information isn't actually relevant. Its always good that people who are wrong get a cogent reply, but beyond that adding a "me too" post doesn't have any more benefit than a downvote does. I used to just upvote post I agreed with, but since those are no longer public the only way to clue readers in to what the consensus…

I'm not sure I'd agree he was arguing for the irrelevance of anecdotal evidence, only that it has a disproportionate impact when considered with statistical evidence. Rather than censoring the anecodotes, the better solution is to simply point out the danger of overweighting them.

The purpose of a site like HN is not to arrive at some kind of imaginary consensus, it's to inform and engage people in meaningful discussion. There is no winning side to be on, and no ultimate arbiter of truth. By downvoting as you do, you rob the site of content in exchange for an illusory sense of victory.

Re: You Should Downvote Contrarian Anecdotes

#145

Post by brudgers »

Symmetry wrote: post #134
brudgers wrote: post #58

By this logic, HN'ers should downvote any anecdote describing successfully exiting a startup because statistics show that the vast majority of startups fail and these anecdotes describing success unduly influence the thinking of entrepreneurs.

Only if its offered as a counterexample to the proposition that most startups fail.

One might suggest that that such a proposition is an underlying premise of HN.

Re: You Should Downvote Contrarian Anecdotes

#146

Post by guelo »

beagle3 wrote: post #140
guelo wrote: post #129

Earlier quoted context omitted.

It's one thing when learned people dig into the science and find issues to disagree about, it is completely different when some ignorant reporter or blogger disagrees just based on anecdotes or how truthy it feels to them

Someone who is good with statistics is more learned than the prominent experts in many areas when it come to disagreement. See e.g. this: http://blog.sethroberts.net/2012/04/08/gene-linked-to-autism... - Roberts is a professor of psychology, but he wears a statisticians hat in this post. He does this often and in many fields, and more often than not, his review (though well argued) is discarded as "a blogger who disa…

You have provided no support for the idea that MOST scientists don't understand statistics. You link to criticism made about one reporter. I am much more prepared to believe that reporters don't understand statistics than scientists. Still, you have provided an anecdote in support of broad sweeping statements.

Re: You Should Downvote Contrarian Anecdotes

#147

Post by guelo »

retrogradeorbit wrote: post #28

In a world were 47 of 53 landmark cancer research papers were found to not be replicable... where one scientist admitted (after the meta-researchers tried 50 times to replicate the result and failed) that they had run the experiment 6 times and got the result once and reported that because it made a better "story"... where the act of suppressing refutations of other's bogus research is "culturally commonplace"... Som…

Stop using anecdotes to support extremely broad statements about all of science.

Re: You Should Downvote Contrarian Anecdotes

#148

Post by beagle3 »

guelo wrote: post #146
beagle3 wrote: post #140

Earlier quoted context omitted.

Someone who is good with statistics is more learned than the prominent experts in many areas when it come to disagreement. See e.g. this: http://blog.sethroberts.net/2012/04/08/gene-linked-to-autism... - Roberts is a professor of psychology, but he wears a statisticians hat in this post. He does this often and in many fields, and more often than not, his review (though well argued) is discarded as "a blogger who disa…

You have provided no support for the idea that MOST scientists don't understand statistics. You link to criticism made about one reporter. I am much more prepared to believe that reporters don't understand statistics than scientists. Still, you have provided an anecdote in support of broad sweeping statements.

> You have provided no support for the idea that MOST scientists don't understand statistics.

That's true, but neither did you (or anyone else ever, for that matter) provide support for the idea that MOST scientists do understand statistics. See how easy it is to discard anything you disagree with?

> I am much more prepared to believe that reporters don't understand statistics than scientists.

That's fine, but (a) it doesn't say anything about how bad scientists are with statistics (only that they are slightly better than reporters, which I tend to agree with), and (b) this is an argument from bias/faith/religion/prejudice, not from science or data. You are just as guilty as anyone you criticize. You might be more right or less right, but you* don't have the moral ground. (* general you).

> Still, you have provided an anecdote in support of broad sweeping statements.

What was that statement of yours about learned people digging into science? So now it is not enough for those people to know what they are talking about, they have to do it in a format you approve of.

I can provide tens more valid criticisms. I charge $200-$1000/hour for my line of work, and I'd be happy to take as much to work for you finding them, when I have some free time.

But I'll throw in a freebie: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/ - though I suspect it will stop at most people's "no true scotsman/scientist" filter...

Re: You Should Downvote Contrarian Anecdotes

#149

Post by pbreit »

retrogradeorbit wrote: post #24
loup-vaillant wrote: post #23

Earlier quoted context omitted.

Let's take an example, since they seem to work best: vaccines. Vaccines protect you from the risk of contracting particular diseases, some of which are crippling, lethal, or incurable. Plus, most are extremely effective: once you take your shots, you are effectively immune. That's good. There is a downside however. Sometimes, vaccines have side effects. Most side effects are quite benign, but if you're unlucky enough…

Are you actually arguing against giving anecdotes weight by using... anecdotes?

I didn't see any anecdotes in the post.

Re: You Should Downvote Contrarian Anecdotes

#150

Post by lukev »

It's true, anecdotes are not data and have no statistical weight whatsoever.

However, anecdotes can and do (and should) play a large role in influencing how we think. They can humanize a problem and create food for thought in a way that no amount of statistics ever could.

For example, it's one thing to cite statistics that children of gay parents do just as well as those of straight parents (I have no idea if it's true, but it's an interesting contemporary question). But that's not likely to change the mind of a homophobe on the issue of gay adoption. On the flip side, a lucid, heartfelt anecdote from a person who had gay parents might actually help someone to understand what it's like to grow up in that environment an therefore become sympathetic.

Of course, it has no bearing on the actual statistics at all, but sometimes statistics aren't the most important thing.

Post reply on HN