3 ms·

Overplayed? It jeopardises the entire edifice of that subfield, and by extension psychology itself, if it is not vigorously disowned. Sokal said it well (from

by FabHK 5y ago

Overplayed? It jeopardises the entire edifice of that subfield, and by extension psychology itself, if it is not vigorously disowned. Sokal said it well (from the article): > “For me, the real question is not about Fredrickson or Losada or Seligman,” Sokal says. “It’s about the whole community. Why is it that no one before Nick—and I mean Nick was a first semester part-time Master’s student, at, let’s be honest, a fairly obscure university in London who has no particular training in mathematics—why is it that no one realized this stuff was bullshit? Where were all the supposed experts?” > “Is it really true that no one saw through this,” he asks, “in an article that was cited 350 times, in a field which touts itself as being so scientific?”

derbOac 5y ago
350 citations is not much. Retracted studies get cited more than that.
dekhn 5y ago
I think the answer to this is remarkably simple. You can align scientists and fields along a spectrum and group them into roughly two categories: The scientists who make falsifiable theories and run reproducible experiments -and- everybody else. On the left side of the spectrum is physicists working in long-understood areas, such as "predict the motion of elements of the solar system 100 years into the future using Newton's law and observed data", in the middle is "biologists explaining the behavior of C. elegans using a combination of molecular biology and machine learning", and on the far right side is "psychologists attempting to come up with universal rules about human behavior using studies with tiny n" The folks on the far, far left have a fairly straightforward system that works pretty well. If you claim something like "FTL" or "new energy source", you describe your apparatus, publish your data, show people a demo in the lab, and if it's real, somebody else in the world will be motivated to replicate your experiment, or help show you why you're technically wrong. Almost nobody makes it far along this path by repeatedly publishing interesting, but provably wrong results. In that middle zone, there's still an air of replicability, but important studies turn out to be really hard to replicate for dumb reasons, like "the temperature of the lab is off by 0.5C" or "you aren't using the exact same plasmid as mine, which I personally isolated when in <nobel prize winners lab> during my postdoc". It's actually quite easy to publish fairly wrong science, for quite some time, without anybody really noticing. For example, hundreds and hundreds of papers turn out to have copy-pasted gels, where the authors just made up results and published them and their conclusions. I guess nobody bothered to replicate that paper! On the right side is psychologists. I don't even think this field, with the except of a few very limited areas, has the statistical power to make the strong conclusions authors come to, in 99.9% of papers. It's better to think of this area as people just sort of writing what they want to believe, and placing a pseudoscientific veneer on top. This would fall into the "not even wrong" category- it doesn't even really qualify to be evaluated as science. Or if it is, it falls into the category of "speculative complex systems theory".
PeterisP 5y ago
Right after that statement the article describes how people did indeed see through this and quotes three such people.

by FabHK 5y ago

derbOac 5y ago

350 citations is not much. Retracted studies get cited more than that.

dekhn 5y ago

I think the answer to this is remarkably simple. You can align scientists and fields along a spectrum and group them into roughly two categories: The scientists who make falsifiable theories and run reproducible experiments -and- everybody else. On the left side of the spectrum is physicists working in long-understood areas, such as "predict the motion of elements of the solar system 100 years into the future using Newton's law and observed data", in the middle is "biologists explaining the behavior of C. elegans using a combination of molecular biology and machine learning", and on the far right side is "psychologists attempting to come up with universal rules about human behavior using studies with tiny n" The folks on the far, far left have a fairly straightforward system that works pretty well. If you claim something like "FTL" or "new energy source", you describe your apparatus, publish your data, show people a demo in the lab, and if it's real, somebody else in the world will be motivated to replicate your experiment, or help show you why you're technically wrong. Almost nobody makes it far along this path by repeatedly publishing interesting, but provably wrong results. In that middle zone, there's still an air of replicability, but important studies turn out to be really hard to replicate for dumb reasons, like "the temperature of the lab is off by 0.5C" or "you aren't using the exact same plasmid as mine, which I personally isolated when in <nobel prize winners lab> during my postdoc". It's actually quite easy to publish fairly wrong science, for quite some time, without anybody really noticing. For example, hundreds and hundreds of papers turn out to have copy-pasted gels, where the authors just made up results and published them and their conclusions. I guess nobody bothered to replicate that paper! On the right side is psychologists. I don't even think this field, with the except of a few very limited areas, has the statistical power to make the strong conclusions authors come to, in 99.9% of papers. It's better to think of this area as people just sort of writing what they want to believe, and placing a pseudoscientific veneer on top. This would fall into the "not even wrong" category- it doesn't even really qualify to be evaluated as science. Or if it is, it falls into the category of "speculative complex systems theory".

PeterisP 5y ago

Right after that statement the article describes how people did indeed see through this and quotes three such people.