Bad apple spoils eating behavior research
September 21, 2018 12:10 PM Subscribe
Cornell University's “world-renowned eating behavior expert," Brian Wansink, has resigned (WaPo link) following the news, per Vox, that 13 of Wansink’s studies "have now been retracted, including the six pulled from [the Journal of the American Medical Association] Wednesday. Among them: studies suggesting people who grocery shop hungry buy more calories; that preordering lunch can help you choose healthier food; and that serving people out of large bowls encourage[s] them to serve themselves larger portions."
Wansink previously.
Wansink previously.
That's a huge recurring theme with the "replication crisis". Research that doesn't either directly bring income via grants or else raise a researcher's profile (so they can bring income via grants) is much less likely to get done, because not bringing grant money to your institution is career suicide.
posted by tobascodagama at 12:21 PM on September 21, 2018 [14 favorites]
posted by tobascodagama at 12:21 PM on September 21, 2018 [14 favorites]
It’s almost like if we removed the need to compete for grants and fully funded education there wouldn’t be a profit motive to fabricate results like this.
posted by The Whelk at 12:27 PM on September 21, 2018 [84 favorites]
posted by The Whelk at 12:27 PM on September 21, 2018 [84 favorites]
Also, my initial reaction to this story in pop culture terms: "Scientific Progress goes 'Boink'" on steroids
posted by ZeusHumms at 12:27 PM on September 21, 2018 [1 favorite]
posted by ZeusHumms at 12:27 PM on September 21, 2018 [1 favorite]
It’s almost like if we removed the need to compete for grants and fully funded education there wouldn’t be a profit motive to fabricate results like this.
100% this. I'd add that demystifying universities through universal higher ed would do a lot to rebuild public trust of academia, and reintegrate the academy into day-to-day life.
posted by codacorolla at 12:38 PM on September 21, 2018 [23 favorites]
100% this. I'd add that demystifying universities through universal higher ed would do a lot to rebuild public trust of academia, and reintegrate the academy into day-to-day life.
posted by codacorolla at 12:38 PM on September 21, 2018 [23 favorites]
Any word on the landmark study regarding eating 30 minutes before swimming? Asking for concerned parents everywhere.
posted by Abehammerb Lincoln at 12:43 PM on September 21, 2018 [12 favorites]
posted by Abehammerb Lincoln at 12:43 PM on September 21, 2018 [12 favorites]
What does "reintegrate the academy into day-to-day life" mean? There has always been separation between higher education and "day-to-day" life especially once many institutions in the U.S. began trying to adapt the German model of universities that focused exclusively on research and post-graduate study. That tension, in fact, was central in the development of the land-grant university which is arguably the United States's greatest innovation in and contribution to higher education. It was a grand victory by "the grange" (an old-timey word for "farmers and related industry") to require at least one university in each state to include some focus on agriculture and mechanics (hence the "A&M" universities in some states).
posted by ElKevbo at 12:48 PM on September 21, 2018 [7 favorites]
posted by ElKevbo at 12:48 PM on September 21, 2018 [7 favorites]
What does "reintegrate the academy into day-to-day life" mean?
I think it has something to do with the widespread willingness of people to do what they want in the face of accepted facts, like anti-vaxxers, or the way climate scientists are ignored so that people get away with things like doubting climate change just because they don't like the notion of cutting back their own carbon footprint, or reducing the profits of energy companies.
That's anti-intellectualism, right?
posted by wenestvedt at 12:57 PM on September 21, 2018 [4 favorites]
I think it has something to do with the widespread willingness of people to do what they want in the face of accepted facts, like anti-vaxxers, or the way climate scientists are ignored so that people get away with things like doubting climate change just because they don't like the notion of cutting back their own carbon footprint, or reducing the profits of energy companies.
That's anti-intellectualism, right?
posted by wenestvedt at 12:57 PM on September 21, 2018 [4 favorites]
> if we removed the need to compete for grants and fully funded education there wouldn’t be a profit motive
Yes, and also changed publication standards so that null findings from [well-designed, appropriately powered] studies were just as publication-worthy as those with significant p-values -- in short, shift the focus to the design and methodology rather than the result.
posted by Westringia F. at 1:07 PM on September 21, 2018 [42 favorites]
Yes, and also changed publication standards so that null findings from [well-designed, appropriately powered] studies were just as publication-worthy as those with significant p-values -- in short, shift the focus to the design and methodology rather than the result.
posted by Westringia F. at 1:07 PM on September 21, 2018 [42 favorites]
Yes - essentially bringing more people into academic production instead of having it be a reserved category for the elite: either the extremely lucky in the middle and lower class who then feel beholden to bad systems, or the extremely rich who can afford the massive amounts of lost time and productivity that it takes to get a terminal degree and are already inured to playing games of power.
The right is absolutely incorrect in painting academia as a machine of left wing indoctrination, but are absolutely correct in the idea that the way we currently structure higher education and research as being elitist. It's what makes a stooge like Jordan Peterson successful: because his laughable scholarship is enabled by the completely fucked structure of modern academia, which encourages professors to become brands and "thought leaders" to hedge against uncertain job conditions.
The scholar-as-brand isn't really surprising. Situations where individuals like the researcher in the FPP occur are commonplace, because of the environment around them: fighting for what amount to scraps in a hyper-competitive environment that encourage gamesmanship instead of actual knowledge building. I can think of a half-dozen other examples within the past few years that are, thankfully, coming to light because of the replication crisis.
posted by codacorolla at 1:15 PM on September 21, 2018 [10 favorites]
The right is absolutely incorrect in painting academia as a machine of left wing indoctrination, but are absolutely correct in the idea that the way we currently structure higher education and research as being elitist. It's what makes a stooge like Jordan Peterson successful: because his laughable scholarship is enabled by the completely fucked structure of modern academia, which encourages professors to become brands and "thought leaders" to hedge against uncertain job conditions.
The scholar-as-brand isn't really surprising. Situations where individuals like the researcher in the FPP occur are commonplace, because of the environment around them: fighting for what amount to scraps in a hyper-competitive environment that encourage gamesmanship instead of actual knowledge building. I can think of a half-dozen other examples within the past few years that are, thankfully, coming to light because of the replication crisis.
posted by codacorolla at 1:15 PM on September 21, 2018 [10 favorites]
The three studies mentioned in the FPP are of hypotheses so plausible I'd think it would only be worth doing the study in the first place if you had a strong suspicion they weren't true.
How does a researcher become "world-renowned" by telling people that what would probably be their default beliefs if queried are in fact the case?
A professorship would have been wasted on Wansink even if he hadn't committed fraud.
posted by jamjam at 1:29 PM on September 21, 2018 [1 favorite]
How does a researcher become "world-renowned" by telling people that what would probably be their default beliefs if queried are in fact the case?
A professorship would have been wasted on Wansink even if he hadn't committed fraud.
posted by jamjam at 1:29 PM on September 21, 2018 [1 favorite]
jamjam, it's important that we test "default" beliefs because sometimes they're not accurate, and if you build research on top of them without ensuring that the foundation is solid, that research is worthless.
posted by telophase at 1:51 PM on September 21, 2018 [40 favorites]
posted by telophase at 1:51 PM on September 21, 2018 [40 favorites]
I feel like I should start my own scientific journal, Negative Results Monthly, where we publish all the papers that don’t find anything.
Not sure who would subscribe.
posted by Huffy Puffy at 2:03 PM on September 21, 2018 [15 favorites]
Not sure who would subscribe.
posted by Huffy Puffy at 2:03 PM on September 21, 2018 [15 favorites]
A small part of my wife’s job is working on a multi-million dollar, multi-site federally-funded research project. She recently told me how the new PI had basically suggested in a meeting discussing publication ideas to just start running some numbers and see what P-values came up
She put a stop to that idea pretty quick.
posted by Big Al 8000 at 2:12 PM on September 21, 2018 [9 favorites]
She put a stop to that idea pretty quick.
posted by Big Al 8000 at 2:12 PM on September 21, 2018 [9 favorites]
I feel like I should start my own scientific journal, Negative Results Monthly, where we publish all the papers that don’t find anything. Not sure who would subscribe.
Pessimists?
posted by Greg_Ace at 2:28 PM on September 21, 2018 [5 favorites]
Pessimists?
posted by Greg_Ace at 2:28 PM on September 21, 2018 [5 favorites]
Would you say that he was nudged out of his position?
posted by clawsoon at 2:35 PM on September 21, 2018 [2 favorites]
posted by clawsoon at 2:35 PM on September 21, 2018 [2 favorites]
Even without changing the whole competitive system that essentially leads to most scientists having an internal conflict of interest (their skepticism vs their careers), we could improve things a bunch from changing the publishing model:
Right now you get a lot of ‘peer review’ done by barely qualified people who are not paid for the arduous task, and have little incentive to actually put in the large amount of time to do it right.
The big press journals take about $5bn a year from academia (and largely the public), and don’t pay the writers, reviewers, or (most) editors.
Basically, when you don’t pay people, you can’t attract the best talent, but they don’t mind, because they have a nearly captive audience of academic libraries around the word that think they must buy the product, even at very high cost.
It’s a shitty deal for most of the players.
posted by SaltySalticid at 2:39 PM on September 21, 2018 [11 favorites]
Right now you get a lot of ‘peer review’ done by barely qualified people who are not paid for the arduous task, and have little incentive to actually put in the large amount of time to do it right.
The big press journals take about $5bn a year from academia (and largely the public), and don’t pay the writers, reviewers, or (most) editors.
Basically, when you don’t pay people, you can’t attract the best talent, but they don’t mind, because they have a nearly captive audience of academic libraries around the word that think they must buy the product, even at very high cost.
It’s a shitty deal for most of the players.
posted by SaltySalticid at 2:39 PM on September 21, 2018 [11 favorites]
I hope this leads to a whole-scale reassessment of the paradigm where obesity is a psychological/behavioral issue that can be treated with cutesy lifehacks.
posted by Ralston McTodd at 3:01 PM on September 21, 2018 [8 favorites]
posted by Ralston McTodd at 3:01 PM on September 21, 2018 [8 favorites]
Money. Money changes everything.
posted by Mr. Yuck at 3:10 PM on September 21, 2018 [1 favorite]
posted by Mr. Yuck at 3:10 PM on September 21, 2018 [1 favorite]
Who would've thought that an academic centre named the "Food and Brand Lab" would have turned out this way...
posted by danhon at 3:28 PM on September 21, 2018 [1 favorite]
posted by danhon at 3:28 PM on September 21, 2018 [1 favorite]
This reminds me of the Diederik Stapel scandal: Meat eaters more selfish than vegetarians.
posted by Pendragon at 3:44 PM on September 21, 2018
posted by Pendragon at 3:44 PM on September 21, 2018
A difference with Stapel is that Wansink seemed to think that he was doing the right thing and giving good, ethical advice to his grad students. He sincerely thought that p-hacking was good science.
I remember something he wrote where he said that his own grad advisors had said something to the effect that [I'm misremembering a little here, I'm sure] he was producing the lowest-quality work they had ever seen, but he triumphed in the end via hard work and try, try again. He wanted to pass that message on to his own students: Try, try again (with the same data) until you get significant results. Then publish and build your CV. To him, it was an inspiring tale.
posted by clawsoon at 4:05 PM on September 21, 2018 [5 favorites]
I remember something he wrote where he said that his own grad advisors had said something to the effect that [I'm misremembering a little here, I'm sure] he was producing the lowest-quality work they had ever seen, but he triumphed in the end via hard work and try, try again. He wanted to pass that message on to his own students: Try, try again (with the same data) until you get significant results. Then publish and build your CV. To him, it was an inspiring tale.
posted by clawsoon at 4:05 PM on September 21, 2018 [5 favorites]
Huffy Puffy: I feel like I should start my own scientific journal, Negative Results Monthly, where we publish all the papers that don’t find anything. Not sure who would subscribe.
The Journal of Negative Results in Biomedicine was shut down in 2017. You might say that... the result was negative.
(They claim that the result was positive, as "many other journals followed JNRBM’s lead in publishing articles reporting negative or null results. As such JNRBM has succeeded in its mission and there is no longer a need for a specific journal to host these null results.")
posted by clawsoon at 4:10 PM on September 21, 2018 [4 favorites]
The Journal of Negative Results in Biomedicine was shut down in 2017. You might say that... the result was negative.
(They claim that the result was positive, as "many other journals followed JNRBM’s lead in publishing articles reporting negative or null results. As such JNRBM has succeeded in its mission and there is no longer a need for a specific journal to host these null results.")
posted by clawsoon at 4:10 PM on September 21, 2018 [4 favorites]
Reading the discussion on the Wansink blog post that started all this is revealing of how much Wansink just didn't get it. And the quotes in the Vox piece suggest that he still doesn't get it:
posted by clawsoon at 4:27 PM on September 21, 2018 [6 favorites]
“There was no fraud, no intentional misreporting, no plagiarism, or no misappropriation,” he wrote. “I believe all of my findings will be either supported, extended, or modified by other research groups.”Maybe some of his research findings will be replicated. It's possible. Maybe he did none of the bad things he says he didn't do. But he still doesn't get that he was doing bad science, and teaching his students bad science, and that this is bad.
posted by clawsoon at 4:27 PM on September 21, 2018 [6 favorites]
This isn’t just about academic credits. He had more than one bestselling book and probably a lucrative consulting practice.
posted by bq at 4:44 PM on September 21, 2018 [3 favorites]
posted by bq at 4:44 PM on September 21, 2018 [3 favorites]
This guy's research is directly related to this post from two days ago -- the Huffington Post article on weight, doctors and our poisonous culture around body size. His life's work has been to profit from, and intensify, our society's fear and judgment of larger bodies.
Are you fat? It's because you're eating with a big plate, and shopping hungry, and not virtuously planning out your meals like you should! Just follow these 7 simple steps and eat on artificially tiny plates, and fast, and buy new dishes... and if you can't follow them, why? They're so simple. It's a moral failing on your part. There must be something wrong with you. As someone who has only recently pulled herself out of diet culture, this guy has personally contributed to my food neuroses. And apparently he did it with a joyful dismissal of scientific standards and practices. Great.
posted by rogerrogerwhatsyourrvectorvicto at 4:47 PM on September 21, 2018 [24 favorites]
Are you fat? It's because you're eating with a big plate, and shopping hungry, and not virtuously planning out your meals like you should! Just follow these 7 simple steps and eat on artificially tiny plates, and fast, and buy new dishes... and if you can't follow them, why? They're so simple. It's a moral failing on your part. There must be something wrong with you. As someone who has only recently pulled herself out of diet culture, this guy has personally contributed to my food neuroses. And apparently he did it with a joyful dismissal of scientific standards and practices. Great.
posted by rogerrogerwhatsyourrvectorvicto at 4:47 PM on September 21, 2018 [24 favorites]
A day after the resignation a report by Cornell was to be released. I've not found the link. Anyone have it?
posted by soylent00FF00 at 7:23 PM on September 21, 2018
posted by soylent00FF00 at 7:23 PM on September 21, 2018
The Journal of Negative Results in Biomedicine was shut down in 2017. You might say that... the result was negative.
Well, maybe.
posted by Huffy Puffy at 7:34 PM on September 21, 2018 [1 favorite]
Well, maybe.
posted by Huffy Puffy at 7:34 PM on September 21, 2018 [1 favorite]
The internet changed a lot about how knowledge discovery and dissemination works, and I think our scientific and journalistic standards and processes haven’t caught up yet. There are competitive pressures that keep even the most well-intentioned of professionals from being responsible in their work.
Hopefully the eventual resolution of things like the “replication crisis”, the Trump era of unabashedly making false statements, and “fake news”/lack of critical thinking in social media, is that people start getting more interested in learning from really credible sources.
I’d like to see the birth of a new paradigm of responsible science, journalism and public participation. I’d like everyone to have near-complete access to primary sources (e.g. the sources behind journalists’ stories, the data behind scientists’ papers) so that people downstream can double-check claims. The whole “trust me, I’m an expert” thing wears thin when experts start abusing their power. Then you get members of the public throwing the baby out with the bath water, and not trusting even the people who are trying to be responsible scientists/journalists. And ironically, falling for the claims of people who seek to manipulate them by using advances in psychology to target them (see: election interference through targeted advertising). I think the solution to the whole thing is show your data.
Note: there has historically been an argument for journalists keeping some sources confidential, but I see that being abused quite a bit nowadays. E.g. the NYT’s habit of quoting anonymous sources within the Trump WH administration. You can shape any kind of story you want with practices like that. The default should be people standing behind claims that are put into the public sphere, so that there can be follow-up discussion and the opportunity to dig deeper.
Schoolchildren should be taught to expect to be able to follow any claim back to its primary sources. That’s how you prevent gossip, hearsay, etc. It’s so much easier now that we have 24/7 access to the sum total of human knowledge.
The first step is to have a place where this primary data gets housed, and a way to protect it from manipulation (intentional or inadvertent). Maybe this is an actual application of value for blockchain? Next, people build analyses (mathematical, scientific, economic, etc) on top of the data. It’s all transparent, a la open-source, so that anyone can participate in the process, and any assumptions made during the analysis are fully visible. There would be peer reviewing, but this would also be transparent, so that it would be evident if someone was acting in bad faith.
Who would run the whole thing and how you would keep them from being corrupted, I do not know. But you can see elements of these ideas already on Google, Wikipedia, Stackoverflow, Github, Quora, Kaggle. Maybe the UN model with its balance of power also has something to teach us. I’m also not sure how this vision would allow for professionals to make a living from doing their work when so much is put out in the open, but some smart person will figure out a model that works, and then we’ll think back on this as a dark age of sorts. “I can’t believe people used to take scientists’ claims as gospel when they hadn’t published full documentation of their data and methods!”
posted by mantecol at 11:26 PM on September 21, 2018 [1 favorite]
Hopefully the eventual resolution of things like the “replication crisis”, the Trump era of unabashedly making false statements, and “fake news”/lack of critical thinking in social media, is that people start getting more interested in learning from really credible sources.
I’d like to see the birth of a new paradigm of responsible science, journalism and public participation. I’d like everyone to have near-complete access to primary sources (e.g. the sources behind journalists’ stories, the data behind scientists’ papers) so that people downstream can double-check claims. The whole “trust me, I’m an expert” thing wears thin when experts start abusing their power. Then you get members of the public throwing the baby out with the bath water, and not trusting even the people who are trying to be responsible scientists/journalists. And ironically, falling for the claims of people who seek to manipulate them by using advances in psychology to target them (see: election interference through targeted advertising). I think the solution to the whole thing is show your data.
Note: there has historically been an argument for journalists keeping some sources confidential, but I see that being abused quite a bit nowadays. E.g. the NYT’s habit of quoting anonymous sources within the Trump WH administration. You can shape any kind of story you want with practices like that. The default should be people standing behind claims that are put into the public sphere, so that there can be follow-up discussion and the opportunity to dig deeper.
Schoolchildren should be taught to expect to be able to follow any claim back to its primary sources. That’s how you prevent gossip, hearsay, etc. It’s so much easier now that we have 24/7 access to the sum total of human knowledge.
The first step is to have a place where this primary data gets housed, and a way to protect it from manipulation (intentional or inadvertent). Maybe this is an actual application of value for blockchain? Next, people build analyses (mathematical, scientific, economic, etc) on top of the data. It’s all transparent, a la open-source, so that anyone can participate in the process, and any assumptions made during the analysis are fully visible. There would be peer reviewing, but this would also be transparent, so that it would be evident if someone was acting in bad faith.
Who would run the whole thing and how you would keep them from being corrupted, I do not know. But you can see elements of these ideas already on Google, Wikipedia, Stackoverflow, Github, Quora, Kaggle. Maybe the UN model with its balance of power also has something to teach us. I’m also not sure how this vision would allow for professionals to make a living from doing their work when so much is put out in the open, but some smart person will figure out a model that works, and then we’ll think back on this as a dark age of sorts. “I can’t believe people used to take scientists’ claims as gospel when they hadn’t published full documentation of their data and methods!”
posted by mantecol at 11:26 PM on September 21, 2018 [1 favorite]
"I hope this leads to a whole-scale reassessment of the paradigm where obesity is a psychological/behavioral issue that can be treated with cutesy lifehacks."
The idea that eating more leads to obesity didn't hinge on this new guy's body of work. He was using the other existing studies as a haystack when he was trying to hide his needle.
posted by Selena777 at 7:47 AM on September 22, 2018
The idea that eating more leads to obesity didn't hinge on this new guy's body of work. He was using the other existing studies as a haystack when he was trying to hide his needle.
posted by Selena777 at 7:47 AM on September 22, 2018
mantecol, I"m in general agreement.
The one thing is that scientists need to be public about the computer programs they use to analyse their data, not just the data.
posted by Nancy Lebovitz at 9:56 AM on September 22, 2018 [1 favorite]
The one thing is that scientists need to be public about the computer programs they use to analyse their data, not just the data.
posted by Nancy Lebovitz at 9:56 AM on September 22, 2018 [1 favorite]
Hooray I can eat out of a giant bowl now
(I kid, I kid)
posted by batter_my_heart at 2:31 PM on September 22, 2018
(I kid, I kid)
posted by batter_my_heart at 2:31 PM on September 22, 2018
It’s almost like if we removed the need to compete for grants and fully funded education there wouldn’t be a profit motive to fabricate results like this.
There is even more weirdness in the American grant system for psych research. Much of seems to be based on personal relationships and officers at the granting institution who help 'shepherd' the grant through the process. This help can range from advice on application targeting to actual internal advocacy and campaigning. It reinforces some of the systemic inequality between research institutions because alumni and friend networks can play a very strong part in what science gets funded. It also puts a lot of weight of rather unscientific weight on researcher charisma and pushiness over good science.
posted by srboisvert at 5:18 AM on September 23, 2018 [1 favorite]
There is even more weirdness in the American grant system for psych research. Much of seems to be based on personal relationships and officers at the granting institution who help 'shepherd' the grant through the process. This help can range from advice on application targeting to actual internal advocacy and campaigning. It reinforces some of the systemic inequality between research institutions because alumni and friend networks can play a very strong part in what science gets funded. It also puts a lot of weight of rather unscientific weight on researcher charisma and pushiness over good science.
posted by srboisvert at 5:18 AM on September 23, 2018 [1 favorite]
A small part of my wife’s job is working on a multi-million dollar, multi-site federally-funded research project. She recently told me how the new PI had basically suggested in a meeting discussing publication ideas to just start running some numbers and see what P-values came up
She put a stop to that idea pretty quick.
There are ways that this can be legit. Sequential Analysis is where you plan ahead of time to peek at your results and you control your alpha - from clinical trials where they may want to abort early if it looks useless or harmful. However, you must plan this in advance.
posted by srboisvert at 5:27 AM on September 23, 2018
She put a stop to that idea pretty quick.
There are ways that this can be legit. Sequential Analysis is where you plan ahead of time to peek at your results and you control your alpha - from clinical trials where they may want to abort early if it looks useless or harmful. However, you must plan this in advance.
posted by srboisvert at 5:27 AM on September 23, 2018
I assume - and correct me if I'm wrong - that searching through data for correlations might be a good way to get ideas for future studies. "Hey, in this data men are eating more in the presence of women. We should do a proper study to find out if that's true."
posted by clawsoon at 5:39 AM on September 23, 2018
posted by clawsoon at 5:39 AM on September 23, 2018
I assume - and correct me if I'm wrong - that searching through data for correlations might be a good way to get ideas for future studies. "Hey, in this data men are eating more in the presence of women. We should do a proper study to find out if that's true."
Almost.
The really right thing to do would be to say "Hey, in these data men are eating more in the presence of women. I wonder if that's because of Reason X to do with Larger Theory Y. We should do a proper study to see if we can see things that are consistent with Reason X." This might be just doing more of the same thing to see whether the same correlation persists, or might be survey-based research looking for responses consistent or inconsistent with Reason X, or might mean putting men in controlled meals with and without women and drawing blood to check for chemistry markers consistent with Reason X, and so on.
posted by GCU Sweet and Full of Grace at 6:14 AM on September 23, 2018
Almost.
The really right thing to do would be to say "Hey, in these data men are eating more in the presence of women. I wonder if that's because of Reason X to do with Larger Theory Y. We should do a proper study to see if we can see things that are consistent with Reason X." This might be just doing more of the same thing to see whether the same correlation persists, or might be survey-based research looking for responses consistent or inconsistent with Reason X, or might mean putting men in controlled meals with and without women and drawing blood to check for chemistry markers consistent with Reason X, and so on.
posted by GCU Sweet and Full of Grace at 6:14 AM on September 23, 2018
Cornell is letting him retire at the end of the academic year?! The salary they're going to keep paying him to do nothing is likely enough to pay off several times over the student loans I took out for grad school there more than two decades ago, loans that I'm still struggling to pay.
posted by mareli at 6:23 AM on September 23, 2018 [1 favorite]
posted by mareli at 6:23 AM on September 23, 2018 [1 favorite]
The internet changed a lot about how knowledge discovery and dissemination works, and I think our scientific and journalistic standards and processes haven’t caught up yet. There are competitive pressures that keep even the most well-intentioned of professionals from being responsible in their work.
I may be reading this wrong, but I think it's worth pointing out that two of the stories that broke most publicly over the past two years were from the mid-20th century: the Prison Experiment and The Marshmallow Test. Both suffer from the same professor-as-brand problem, and both were pushed and amplified through traditional top down media.
Otherwise, I think I mostly agree with your comment. Two caveats would be:
> Human focused work with a greater than normal risk of harm if the data is released. Dis-aggregating and identifying individual participants in a data set is definitely possible - especially if you can use a bit of snooping to figure out which "Large metropolitan area in the American midwest," that author at the University of Chicago could possibly be referring to.
> Qualitative work has had 'show as much data as possible' as a major determinate of rigor for years, which is often difficult with conference paper and journal article page limits that favor quantitative work. Typically you are limited in how much direct data and rich description you can put into something, because 40 years ago someone at Elsevier realized that they make the most money publishing (now irrelevant) print copies of their journals if each article is 8,000 words long. In my own work, I would be fine publishing full-length interview protocols as appendices, although again there's a human subjects concern in putting that up without careful editing.
posted by codacorolla at 8:04 AM on September 23, 2018 [2 favorites]
I may be reading this wrong, but I think it's worth pointing out that two of the stories that broke most publicly over the past two years were from the mid-20th century: the Prison Experiment and The Marshmallow Test. Both suffer from the same professor-as-brand problem, and both were pushed and amplified through traditional top down media.
Otherwise, I think I mostly agree with your comment. Two caveats would be:
> Human focused work with a greater than normal risk of harm if the data is released. Dis-aggregating and identifying individual participants in a data set is definitely possible - especially if you can use a bit of snooping to figure out which "Large metropolitan area in the American midwest," that author at the University of Chicago could possibly be referring to.
> Qualitative work has had 'show as much data as possible' as a major determinate of rigor for years, which is often difficult with conference paper and journal article page limits that favor quantitative work. Typically you are limited in how much direct data and rich description you can put into something, because 40 years ago someone at Elsevier realized that they make the most money publishing (now irrelevant) print copies of their journals if each article is 8,000 words long. In my own work, I would be fine publishing full-length interview protocols as appendices, although again there's a human subjects concern in putting that up without careful editing.
posted by codacorolla at 8:04 AM on September 23, 2018 [2 favorites]
« Older 15 dollars per ounce | THIS PLACE BLOWS I QUIT Newer »
This thread has been archived and is closed to new comments
posted by ZeusHumms at 12:14 PM on September 21, 2018 [8 favorites]