Anonymous #1 8w

i read it more as like a participation trophy for people who couldn’t hack actual science. (i don’t wholly believe the social sciences aren’t science but if the majority of published literature in your field of ‘science’ can’t be reproduced, and was published nonetheless, that’s not science)

0

Anonymous #2 8w

show me the reproducibility crisis in physics.

1

Anonymous OP 8w

Not exactly physics but you’ll see what I’m getting at. The NIST World Trade Center 7 report compared to the University of Alaska Fairbanks report, you can take the same data and make wildly different conclusions

1

Anonymous #2 8w

i’m sorry, your example is root cause analysis of the world trade center collapse? the example you chose to show reproducibility issues in physics is a postmortem case study on a building collapsing after it got hit by a plane? First: that’s engineering forensics, not physics, second: case studies are not really what the reproducibility centers on, and third: that’s fucking 9/11.

1

Anonymous OP 8w

The reproducibility crisis doesn’t just apply to experimental research, it applies to any research conducted with data, so like… all of science

1

Anonymous #2 8w

Then show it to me in physics, which is very much a branch of science.

1

Anonymous OP 8w

Do you think engineering is just completely separated from physics? Running computer models is running an experiment, obviously we can’t go building up and demolishing buildings for science (all the time) so we use the best models we can find. Even basic solidworks licenses can do some pretty crazy things that stop a lot of problems before the first shovel hits the ground

1

Anonymous #2 8w

I would answer your asinine question but like cmon dude at least pretend like we’re in college

1

Anonymous #2 8w

Brother my undergrad degree is in physics and I’m doing a fucking masters in engineering. Physics research and engineering research/reports are two entirely separate aspects of academia. The base content is related, but the research goals are entirely different.

1

Anonymous #2 8w

The root problem is the unique human ability to take objective fact and conjure your own conclusions about them

1

Anonymous #2 8w

No it’s not. The reproducibility crisis arises from a majority of foundational publications in psychology relying on experiments that can’t be reproduced because of small sample sizes, p hacking, and publication bias. Physics research does not experience this problem because physics research involves direct measurement of physical reality, which is largely unchanging at most scales. The double slit experiment is immune to the reproducibility crisis because it is a result of the physical world.

1

Anonymous OP 8w

My whole argument is that it applies to all the sciences, physics, engineering, biology, political science, even history it doesn’t matter. We’ve built a system where companies pay for research that would otherwise never happen but so many results are unable to be reproduced because the sponsor didn’t sponsor for the truth, just to back up whatever claims they wanted to make

1

Anonymous #2 8w

It’s more common in others and less in some but the root issue still exists across all the sciences

1

Anonymous #2 8w

What’s your connection to physics, genuinely? Why are you trying to explain physics academia to someone with a physics degree?

1

Anonymous OP 8w

Again, more common in others less common in some. Psychology can be heavily influenced by capital because of the pharmacy industry or even companies that want to prove their interpretation of “being more productive” is actually better for your health. Physics maybe not so much because there isn’t some profit motive lying in physics

1

Anonymous #2 8w

Here, an article from the University of Pittsburgh specifically highlighting the inability to reproduce a quantum computing experiment done by Microsoft who then retracted the paper. Is that physics enough?

1

Anonymous #2 8w

No, it’s because of the culture of physics academia. In physics results are treated with cynicism and reproduction in other labs is a key step in the process of discourse/acceptance into the general canon of literature. Physicists immediately try to reproduce novel results, which means that non reproducible results are quickly refuted.

1

Anonymous OP 8w

Are you suggesting that prompt peer review doesn’t happen in other sciences?

1

Anonymous #2 8w

Yes, the Majorana shit is a recent instance of the system fucking working, you dolt. The reproducibility crisis in other fields arises from historically significant publications that shaped textbooks and understanding being unreproducible years later. the initial majorana publication was like a year ago, and it’s been retracted BECAUSE it couldn’t be reproduced

1

Anonymous #2 8w

I’m saying that the norm in physics academia is to attempt to reproduce novel results, to verify them. I won’t attribute a particular cause, but there is some reason that much of the published academic canon of Psychology (like decades old foundational papers) has been found irreproducible. Majorana was a cutting edge paper that was retracted three years after publication.

1

Anonymous OP 8w

Majorana is evidence that Physics can weed out bad results. In psychology the replication crisis resulted in the reevaluation of papers as long as 16 years after publication, as well as several 20 to 40 year old textbook phenomena. That is fundamentally different than result involving a tech giants prototype quantum chips being retracted in three years.

1

Anonymous OP 8w

That’s just the advancement of research? Physics and psychology are two completely different fields that advance at different rates. What may have been foundational to psychology 20-40 years ago may be found to not be true with better technology and understanding, just like physics advanced rapidly during the late 18th and 19th century. The reproducibility crisis is explicitly in reference to the tendency of papers sponsored by major companies to skew research towards a result they prefer

1

Anonymous #2 8w

Perhaps we’re talking about two different things with similar titles, but to me it seems like you’re describing research and not necessarily a reproducibility crisis

1

Anonymous #2 8w

That’s not what it is w/r/t psych. Physics is fundamentally different from psychology. You do not understand how physics research is conducted and interpreted.

1

Anonymous #2 8w

There’s been a particular problem in psychology where many foundational results that are taught in schools have been found to be irreplicable as a result of small sample sizes, p hacking, and publication bias. Research and statistical analysis was being performed sloppily, which resulted in bad papers being published. People attempted to reproduce the results given many experiments had become easier to conduct, and found that many results fully couldn’t be replicated.

1