Psychology education, Philip Zimbardo (the Stanford prison experiment fraud), and the sorry state of our educational system
The semester is over. I'm taking a break from impeachment-mania to write something that just needs to be written.
Look, folks. Y'all know I just detest liars. They... trigger me. Liars in academia? Oh, yeah. They are all over the place, they are a big problem, and something must be done.
Hopefully, you know by now that "the Stanford prison experiment" was a fraud. Here's a quick write-up from Vox, which even addresses education. The "experiment" was staged and scripted. Philip Zimbardo, the "scholar" behind the con, was unmasked as a con artist almost two years ago. How did Zimbardo keep the con going for so long? Simple. The premise of the con was that the "prison" he set up among his students at Stanford deteriorated into abusive behavior so quickly that the "experiment" had to be shut down almost immediately. What does that mean? It means that any attempt to replicate it would be rejected by any other professor's IRB for potential harm to human subjects. It was an insidious con job. He got to be a big star, and nobody could even try to replicate his "study" for the very same reason that he became a star!
And it got written into our educational system. Deeply. Why am I referencing it now?
This semester, I have been teaching a seminar called "Social Science & Science Fiction." Cool class. The first book I had students read was Charles Stross's Glasshouse. Stross published the book in 2005, years before Zimbardo was unmasked, and it was heavily influenced by the Zimbardo con. When I first read the book, I thought Zimbardo was on the level. It's still a really good book, and there is so much in it, from a social science perspective, to make it worth using as a pedagogic tool, but by putting it on the syllabus, I have to/get to rant about the Zimbardo con.
But here's the thing. I had a bunch of freshmen coming into the class, having taken AP Psychology last year, being taught Zimbardo as gospel truth. They had no idea that it was fraud. Their jaws dropped, expressions of horror, anger at their high school teachers, the whole system...
And you know what? Do the math on this. They took those AP Psych classes their senior years of high school. 2018-9. They come into my class to be directed to the debunkery, in 2019. Their high school Psychology classes-- AP Psychology-- were teaching them Zimbardo after he had been unmasked as a fraud in early 2018!
Those students were pissed. And rightfully so.
It happened again in my Congress class, towards the end of the semester. For no vitally important reason, I made a passing reference to the fact that Zimbardo was a fraud. Another freshman, AP Psychology, last year. Horror and anger. The upperclassmen at least had the excuse that when they took high school Psychology, it was pre-debunking, but still.
OK, you know, at some level, AP curriculum takes time to revise. If the test is going to test them on Zimbardo, the class has to expose them to Zimbardo, and there will be a lag period. Zimbardo was unmasked in early 2018, and maybe there just wasn't time to revise the AP test.
But...
And as they say, "everything before the 'but' is irrelevant..."
Those "AP" teachers should have bloody well known that Zimbardo was debunked and unmasked as a fraud. They should read. You know, "reading?" That thing you're doing now?
They are teaching students that a major fraud was a legitimate study, after the "study" was unmasked as fraud.
And I want to see what happens next year. Will next year's incoming freshmen still be shocked? Will they still have been taught Zimbardo as gospel truth? How long will it take these people to get their act together and stop teaching the Stanford prison fraud as though it had been an actual experiment rather than a con on the entirety of academia?
The credibility of our scientific education matters. In order to have that credibility, the educational system needs to respond to fraud by adjusting its instruction. So far, it's not doing that. That is to the detriment of every form of scientific instruction because it undercuts the credibility of every scientific endeavor.
Our schools are teaching fraud as truth.
And it continues to the collegiate level. I ain't namin' names here, but I do know of academics-- professors-- who still teach Zimbardo without explaining that he was a fraud. Look, this isn't like Einstein denying quantum mechanics because he was an old-school physicist, yielding Max Planck's great line, science advances one funeral at a time. There's straight-up denialism going on from people who either don't follow the news, or don't want to deal with the fact that a study so critical to how they view their discipline was fraud.
We cannot have any credibility in scientific education if our educational system teaches fraudulent work as though it had been a real study.
No. You can't do that. You do that in Psychology and you undercut every scientific field.
Hey, kids! Never trust your teachers. Never trust your professors. Don't trust me! Check everything. It would be nice if that didn't have to be the case, but it does. There's fraud in the system, and your teachers are teaching it to you, even after it has been debunked. Behold! The secondary and post-secondary educational system of the U.S.!
And yes, Psychology has taken the worst of it.
Psychology, right now, is still in the grips of a "replication crisis," as I have to remind/inform people on a regular basis. A bunch of high profile and foundational Psychology studies are just bunk, but treated as gospel truth because for periods of time that can range up to decades, nobody has bothered trying to replicate them. And in Zimbardo's case, the problem was that nobody could. Science is about replication. If your results can't be replicated, they're nothing, and Psychology is filled with un-replicable results. That kind of means Psychology has some major problems.
To its credit, Psychology is trying to address that. The "replication crisis" exists because people in the discipline are now actually trying to replicate their findings. That's... science! This is the corrective process! I frequently sound like I'm bashing Psychology as a discipline, but what's going on right now is the necessary corrective process for past problems. This is good! It's hard, but it's good. The point of this post is to bash how Psychology is taught because once the debunking has happened, it doesn't seem to be making its way into the curriculum. That's my issue. The "replication crisis" exists because Psychology as a field is going back and doing the right things, as a proper scientific discipline should, in order to correct for past problems. That should give you some skepticism for results that have not yet been subjected to replication, but that's true for all of science. It's just that the educational system is failing here.
To be sure, every discipline has its share of un-replicable results. Bluntly speaking, though, my style of Political Science does pretty well. Quantitative work. Why? Simple. My kind of Political Science revolves around quantitative analysis of publicly available data and surveys that get repeated year after year, like the National Election Studies survey.
Defrauding the field of electoral analysis would be pretty hard. The American National Election Studies survey is done every election cycle. Even if you could insert fraud into one year, it would stick out like a sore thumb. Then, there's the publicly available data, like election results, legislative behavior, and all that stuff. Fraud? Hard here. Not impossible. See, for example, the scandal around Michael LaCour's fabricated data for changing peoples' minds on gay marriage. That was uncovered by a fellow Berkeley Ph.D.-- a total badass named David Broockman. But... it didn't take long for Broockman to pull his social-scientific superheroics, when LaCour made up his own data set.
And there are errors in my style of social science. My kind of social science needs to own up to the Rogoff-Reinhardt Excel spreadsheet typo, which led to the claim that when the debt-to-GDP ratio of a country tops .9, badness ensues. That nonsense took over the world of policy discussion during the financial crisis, and it was a typo in an Excel spreadsheet, uncovered by a UMass Amherst grad student doing... a replication paper for a graduate econometrics course. Credit to Thomas Herndon!
Debunkers are heroes!
So yes, there's some fraud, and some sloppiness. My side is not blame-free. But my side of Political Science gets a lot of replication. Someone easily bored-- let's call him "Justin Buchler"-- might even yawn once or twice about the lack of originality in so much scholarship out there, which beats the same few points into the ground, over and over again, but hey. That's replication. Science! It's why we haven't been as hard-hit by a replication crisis.
And as loathe as some in Political Science are to admit it, qualitative work is more prone to replication problems. And fraud. Yes, it's easier to falsify interview data, or other qualitative work. And when you're in small-n territory, with interpretation being a thing, replication winds up being an issue. Yes, qualitative Political Science is harder to replicate. Some practitioners get really defensive about it, but it's just the truth.
Those who are a little more intellectually honest veer towards Clifford Geertz. Geertz was an anthropologist who preached the value of "thick description," and against any attempt at generalization. He was, bluntly, not a scientist. He was really an anti-scientist, when it came to studies of people, seeing the social world in such a way that a) anything generalizable is not worth studying, and b) your interpretation is just as valid is mine, no matter what you see or what I see anyway. Anti-science and intrinsically un-replicable.
If that's your thing, then OK, but you're not a scientist. In fact, you are rejecting science, and you don't get to appeal to science. If you want to do science, then you need to put things up for replication, and you need to conduct your analysis in a way that is amenable to replication.
And when it doesn't pan out, you need to bloody well stop teaching it.
From now on, I'm going to begin each semester by asking how many of my students know that Zimbardo was unmasked as a fraud. I want to know how long it's going to take for this to percolate down through the educational system. If it will...
Look, folks. Y'all know I just detest liars. They... trigger me. Liars in academia? Oh, yeah. They are all over the place, they are a big problem, and something must be done.
Hopefully, you know by now that "the Stanford prison experiment" was a fraud. Here's a quick write-up from Vox, which even addresses education. The "experiment" was staged and scripted. Philip Zimbardo, the "scholar" behind the con, was unmasked as a con artist almost two years ago. How did Zimbardo keep the con going for so long? Simple. The premise of the con was that the "prison" he set up among his students at Stanford deteriorated into abusive behavior so quickly that the "experiment" had to be shut down almost immediately. What does that mean? It means that any attempt to replicate it would be rejected by any other professor's IRB for potential harm to human subjects. It was an insidious con job. He got to be a big star, and nobody could even try to replicate his "study" for the very same reason that he became a star!
And it got written into our educational system. Deeply. Why am I referencing it now?
This semester, I have been teaching a seminar called "Social Science & Science Fiction." Cool class. The first book I had students read was Charles Stross's Glasshouse. Stross published the book in 2005, years before Zimbardo was unmasked, and it was heavily influenced by the Zimbardo con. When I first read the book, I thought Zimbardo was on the level. It's still a really good book, and there is so much in it, from a social science perspective, to make it worth using as a pedagogic tool, but by putting it on the syllabus, I have to/get to rant about the Zimbardo con.
But here's the thing. I had a bunch of freshmen coming into the class, having taken AP Psychology last year, being taught Zimbardo as gospel truth. They had no idea that it was fraud. Their jaws dropped, expressions of horror, anger at their high school teachers, the whole system...
And you know what? Do the math on this. They took those AP Psych classes their senior years of high school. 2018-9. They come into my class to be directed to the debunkery, in 2019. Their high school Psychology classes-- AP Psychology-- were teaching them Zimbardo after he had been unmasked as a fraud in early 2018!
Those students were pissed. And rightfully so.
It happened again in my Congress class, towards the end of the semester. For no vitally important reason, I made a passing reference to the fact that Zimbardo was a fraud. Another freshman, AP Psychology, last year. Horror and anger. The upperclassmen at least had the excuse that when they took high school Psychology, it was pre-debunking, but still.
OK, you know, at some level, AP curriculum takes time to revise. If the test is going to test them on Zimbardo, the class has to expose them to Zimbardo, and there will be a lag period. Zimbardo was unmasked in early 2018, and maybe there just wasn't time to revise the AP test.
But...
And as they say, "everything before the 'but' is irrelevant..."
Those "AP" teachers should have bloody well known that Zimbardo was debunked and unmasked as a fraud. They should read. You know, "reading?" That thing you're doing now?
They are teaching students that a major fraud was a legitimate study, after the "study" was unmasked as fraud.
And I want to see what happens next year. Will next year's incoming freshmen still be shocked? Will they still have been taught Zimbardo as gospel truth? How long will it take these people to get their act together and stop teaching the Stanford prison fraud as though it had been an actual experiment rather than a con on the entirety of academia?
The credibility of our scientific education matters. In order to have that credibility, the educational system needs to respond to fraud by adjusting its instruction. So far, it's not doing that. That is to the detriment of every form of scientific instruction because it undercuts the credibility of every scientific endeavor.
Our schools are teaching fraud as truth.
And it continues to the collegiate level. I ain't namin' names here, but I do know of academics-- professors-- who still teach Zimbardo without explaining that he was a fraud. Look, this isn't like Einstein denying quantum mechanics because he was an old-school physicist, yielding Max Planck's great line, science advances one funeral at a time. There's straight-up denialism going on from people who either don't follow the news, or don't want to deal with the fact that a study so critical to how they view their discipline was fraud.
We cannot have any credibility in scientific education if our educational system teaches fraudulent work as though it had been a real study.
No. You can't do that. You do that in Psychology and you undercut every scientific field.
Hey, kids! Never trust your teachers. Never trust your professors. Don't trust me! Check everything. It would be nice if that didn't have to be the case, but it does. There's fraud in the system, and your teachers are teaching it to you, even after it has been debunked. Behold! The secondary and post-secondary educational system of the U.S.!
And yes, Psychology has taken the worst of it.
Psychology, right now, is still in the grips of a "replication crisis," as I have to remind/inform people on a regular basis. A bunch of high profile and foundational Psychology studies are just bunk, but treated as gospel truth because for periods of time that can range up to decades, nobody has bothered trying to replicate them. And in Zimbardo's case, the problem was that nobody could. Science is about replication. If your results can't be replicated, they're nothing, and Psychology is filled with un-replicable results. That kind of means Psychology has some major problems.
To its credit, Psychology is trying to address that. The "replication crisis" exists because people in the discipline are now actually trying to replicate their findings. That's... science! This is the corrective process! I frequently sound like I'm bashing Psychology as a discipline, but what's going on right now is the necessary corrective process for past problems. This is good! It's hard, but it's good. The point of this post is to bash how Psychology is taught because once the debunking has happened, it doesn't seem to be making its way into the curriculum. That's my issue. The "replication crisis" exists because Psychology as a field is going back and doing the right things, as a proper scientific discipline should, in order to correct for past problems. That should give you some skepticism for results that have not yet been subjected to replication, but that's true for all of science. It's just that the educational system is failing here.
To be sure, every discipline has its share of un-replicable results. Bluntly speaking, though, my style of Political Science does pretty well. Quantitative work. Why? Simple. My kind of Political Science revolves around quantitative analysis of publicly available data and surveys that get repeated year after year, like the National Election Studies survey.
Defrauding the field of electoral analysis would be pretty hard. The American National Election Studies survey is done every election cycle. Even if you could insert fraud into one year, it would stick out like a sore thumb. Then, there's the publicly available data, like election results, legislative behavior, and all that stuff. Fraud? Hard here. Not impossible. See, for example, the scandal around Michael LaCour's fabricated data for changing peoples' minds on gay marriage. That was uncovered by a fellow Berkeley Ph.D.-- a total badass named David Broockman. But... it didn't take long for Broockman to pull his social-scientific superheroics, when LaCour made up his own data set.
And there are errors in my style of social science. My kind of social science needs to own up to the Rogoff-Reinhardt Excel spreadsheet typo, which led to the claim that when the debt-to-GDP ratio of a country tops .9, badness ensues. That nonsense took over the world of policy discussion during the financial crisis, and it was a typo in an Excel spreadsheet, uncovered by a UMass Amherst grad student doing... a replication paper for a graduate econometrics course. Credit to Thomas Herndon!
Debunkers are heroes!
So yes, there's some fraud, and some sloppiness. My side is not blame-free. But my side of Political Science gets a lot of replication. Someone easily bored-- let's call him "Justin Buchler"-- might even yawn once or twice about the lack of originality in so much scholarship out there, which beats the same few points into the ground, over and over again, but hey. That's replication. Science! It's why we haven't been as hard-hit by a replication crisis.
And as loathe as some in Political Science are to admit it, qualitative work is more prone to replication problems. And fraud. Yes, it's easier to falsify interview data, or other qualitative work. And when you're in small-n territory, with interpretation being a thing, replication winds up being an issue. Yes, qualitative Political Science is harder to replicate. Some practitioners get really defensive about it, but it's just the truth.
Those who are a little more intellectually honest veer towards Clifford Geertz. Geertz was an anthropologist who preached the value of "thick description," and against any attempt at generalization. He was, bluntly, not a scientist. He was really an anti-scientist, when it came to studies of people, seeing the social world in such a way that a) anything generalizable is not worth studying, and b) your interpretation is just as valid is mine, no matter what you see or what I see anyway. Anti-science and intrinsically un-replicable.
If that's your thing, then OK, but you're not a scientist. In fact, you are rejecting science, and you don't get to appeal to science. If you want to do science, then you need to put things up for replication, and you need to conduct your analysis in a way that is amenable to replication.
And when it doesn't pan out, you need to bloody well stop teaching it.
From now on, I'm going to begin each semester by asking how many of my students know that Zimbardo was unmasked as a fraud. I want to know how long it's going to take for this to percolate down through the educational system. If it will...
Comments
Post a Comment