Science is really f*cking hard


#1

Originally published at: http://boingboing.net/2015/08/20/science-is-really-fcking-hard.html


#2

I always wondered why the Scientific Method (as I was taught in high school) requires a hypothesis up front. That seems to me to give an incentive to game the study before you’ve even started. The method sets up right away a situation where you are personally vested in an outcome.

I would think there should be an option of “I don’t know what will happen, I’m not trying to prove or disprove anything, but I’ll just study this thing I’m curious about find out where the results lead.”


#3

The idea (at least partially) behind having the hypothesis up front is an effort to provide focus- both to the direction of the research, as well as to the methods used to prove/disprove said thesis. The “curious” method is prone to other types of failure (besides the vested outcome problem).
$0.02


#4

Am I the only one who’s been paying attention all these years? News flash, kids: This has been going on at least since I became old enough to read a newspaper, back in the seventies. I eventually learned that I was about to read some bullshit when I saw the line: “A new study shows [this ordinary food/common drug] [prevents/cures/causes] [some disease, usually cancer].”
It seems that almost any study related to the fields of psychology, the social sciences, or food-as-medicine can be dismissed out of hand. Bogus studies have been used extensively and shamelessly on both sides of the highly politicized tobacco issue. But since skepticism is only reserved for the “bad guys”, the anti-tobacco movement has gotten away with some whoppers. It’s not that there’s a sudden crises, it’s that people have, one hopes, finally begun to wake up and pay attention.


#5

Not having a hypothesis up front is actually one of the big problems that’s only now being addressed. The problem is spurious “statistically significant” results resulting from a kind of shotgun approach in the sense of https://xkcd.com/882/


#6

Science isn’t getting any harder… it’s just that dishonesty is increasingly rewarded more.


#7

Scientific community has been corrupt for decades. Money talks, period. Altruism? Doctors at one time made house calls.

Looking at the graphic it looks like a gel chromatograph. Asking which political party has what effect on the economy it specifically exempts recessions. So which party can create the most outrageous bubbles?


#8

Actually, the green jelly beans contain a dye that the other colors don’t, that may be sensitizing to some consumers. That could explain the correlation. Rerun the experiment and target it to prove or disprove this hypothesis.


#9

After Seismologists were sent to jail in Italy for saying there probably wouldn’t be a major earthquake, just before a major earthquake happened, our guys are very wary of Media.

Plenty of cases where a report asks, “Does this small quake mean a Big quake is coming?” “Well, it’s unlikely but it’s possible” then becomes the obvious headline.

I have a whole lot more faith I the scientific community that almost any other part of society. The bulk of them are nerdish, intelligent enthusiasts in their field. Unusually detached from greed and politics compared to most folks.

Of course you have the Research To Prove Your Bullshit companies. That’s definitely not the majority.


#10

The scientific method you were taught in high school is bullshit. No scientist has ever worked that way.


#11

Ask me some time about plagiarism in scientific papers.


#12

The resource that really helped me in all this is Less Wrong. You may find the author’s transhumanism a little much, but his work on rationality, Bayes’ Theorem and how they all apply in the scientific method is really good. And very down-to-earth language.


#13

The hidden text says “‘So, uh, we did the green study again and got no link. It was probably a–’ ‘RESEARCH CONFLICTED ON GREEN JELLY BEAN/ACNE LINK; MORE STUDY RECOMMENDED!’”

You might also notice that 1 of the twenty experiments produced the spurious correlation, kind of what a p<0.05 represents.


#14

I appreciated the irony (appropriateness? murphy’s lawfulness? TANJ-ness?_ at the end of the article -

CORRECTION (Aug. 19, 12:10 p.m.): An earlier version of the p-hacking interactive in this article mislabeled one of its economic variables. It was GDP, not productivity.

Think they did that on purpose?


#15

This topic was automatically closed after 5 days. New replies are no longer allowed.