Unfortunately, this is science too.
from The_Picard_Maneuver@lemmy.world to science_memes@mander.xyz on 27 May 13:46
https://lemmy.world/post/30334605

#science_memes

threaded - newest

chonomaiwokurae@lemm.ee on 27 May 14:36 next collapse

Proving that something doesn’t work can be valuable data, too. Especially in research close to industrial interests… celebrate failures!

drre@feddit.org on 27 May 15:01 collapse

well yeah, but there is money in knowing what to avoid. in academia it’s more like “why can’t i reproduce this effect i read about in this fancy paper, am i stupid or what”, when maybe, they just got lucky, or had plenty of very reasonable analysis options to choose from, or simply fudged the numbers. i fear that in much of academia there is a huge incentive to publish at whatever cost

LibertyLizard@slrpnk.net on 27 May 14:40 next collapse

Make sure you publish that shit somehow so the next person doesn’t waste their time on the same experiment.

DannyBoy@sh.itjust.works on 27 May 14:56 next collapse

I didn’t know Richard Stallman did research.

rustydrd@sh.itjust.works on 27 May 18:16 collapse

Actually, this is Stannis Baratheon from The Witcher.

miss_demeanour@lemmy.dbzer0.com on 27 May 15:02 next collapse

“Give me six lines of data harvested from the most honourable of men, and I will find an excuse in them to hang him.”

– Pileated Woodpecker Richdude

conditional_soup@lemm.ee on 27 May 15:03 next collapse

I remember listening to an episode of TWiV where they bemoaned that more negative results weren’t published. They’re useful, too, just not nearly as cool and flashy as positive results.

FundMECFSResearch@lemmy.blahaj.zone on 27 May 15:08 next collapse

Big respect to researchers who publish and share statistically insignificant results.

Instead of doing what is far too common in science, manipulating the data until you find “significance” through twisted interpretations.

rikudou@lemmings.world on 27 May 19:50 next collapse

Biology papers and Photoshop, name a more iconic duo.

prex@aussie.zone on 28 May 11:46 collapse

xkcd.com/882/

Probius@sopuli.xyz on 28 May 11:57 collapse

Is it valid science if you re-test the one that had the link to see if it was a fluke?

AHemlocksLie@lemmy.zip on 28 May 15:41 collapse

Not just valid, I’d argue important. It doesn’t make the most exciting headlines and doesn’t get funding very well, though, so it’s not done nearly as often as it should be. A big part of science is not taking things at face value and verifying that there is sufficient proof for claims.

Plus, if both results agree, it statistically tightens the probability of a coincidence. The chances of a 5% chance event happening twice in a row is 0.25%, and three times in a row is 0.0125% so repetition can make the results more certain.

fossilesque@mander.xyz on 27 May 17:35 next collapse

Lol I love this

rustydrd@sh.itjust.works on 27 May 18:14 next collapse

Null results are still results!

friendlymessage@feddit.org on 27 May 21:07 collapse

True, but try to get them published

rustydrd@sh.itjust.works on 28 May 11:27 collapse

I can say with some pride that I have at least co-authored papers with null results, and they did get published. I’m not arguing that what you suggest isn’t true, but I have hope.

peteypete420@sh.itjust.works on 29 May 00:08 collapse

A scholar can never let mere wrongness get in the way of the theory