4 ms·

It's crazy to me that academic fraud isn't a more pressing concern to society in general and academia in particular. The scientific process as currently impleme

It's crazy to me that academic fraud isn't a more pressing concern to society in general and academia in particular. The scientific process as currently implemented is broken across every single discipline. Even subjects like CS that should in theory be trivially reproducible, are rarely so. The reproduction crisis is still going on, but only nerds like us care.

rossdavidh 5y ago
I think there's two problems impeding our ability to focus better on this: 1) for many people, the idea that science has widespread fraud is just hard to accept; in this respect it is similar to the difficulties that many religious communities have in accepting that their clergy could have a corruption problem 2) the solutions require thinking about problems like p-hacking, incentives, selection effects, and other non-trivial concepts that are tough for the average person to wrap their heads around.
- SubiculumCode 5y ago
  Also: I have not seen evidence of widespread fraud. Evidence of fraud,yes. Evidence of widespread fraud no.
  - mistermann 5y ago
    Don't forget though: events proceed evidence, and evidence doesn't always follow events. Also: perception is ~effectively reality.
  - derbOac 5y ago
    That's a fair point, although fraud per se is only a small part of all the problems. There's other forms of corruption than fraud, and a lot of it falls into this zone of plausible deniability rather than outright fraud. Also, I think the problems tend to find most weight with higher concentration of power, so what matters isn't as much "how widespread is corruption?" but rather "how is corruption distributed among power structures in academics and what is rewarded?"
  - rossdavidh 5y ago
    Agreed it's an important point that fraud is only a fraction of the problem.
  - bhk 5y ago
    I have. According to [1], "1 in 4 cancer research papers contains faked data". As the article argues, the standards are perhaps unreasonably strict, but even by more favorable criteria, 1/8 of the papers contained faked data. Interestingly, [2] using the same approach, found fraud in 12.4% of papers in the International Journal of Oncology. More broadly, [2] found fraud in about 4% of the papers studied (782 of 20,621). I'd say that's pretty widespread, but you further have to consider that these papers focused narrowly on a very specific type of fraud that is easy to detect (image duplication), so we would expect the true number of fraudulent papers to be much higher. [1] https://arstechnica.com/science/2015/06/study-claims-1-in-4-cancer-research-papers-contains-faked-data/ https://arstechnica.com/science/2015/06/study-claims-1-in-4-... [2] https://www.biorxiv.org/content/biorxiv/early/2016/04/20/049452.full.pdf https://www.biorxiv.org/content/biorxiv/early/2016/04/20/049...
- derbOac 5y ago
  I've often thought religious corruption is a good analogy, in that many of the societal dynamics are very similar. As I'm writing this the parallels are interesting to think about relative to US politics.
  - throwawayboise 5y ago
    It it is a good analogy. For most lay people, science is a religion. They lack the expertise to understand the theory, but they unquestioningly accept the explanations and interpretations of the so-called experts. Most people don't understand astronomy and physics well enough to prove to themselves that the earth orbits the sun and not vice-versa. Yet they believe it does, with certainty, because they have been taught that it is true.
ArnoVW 5y ago
Aside from reproducibility issues in ML, what sort of issue did you have in mind in CS? Most CS work is 90% maths, I don't see how you can have reproducibility issues?
- the_snooze 5y ago
  Take, for example, network measurement research: https://conferences.sigcomm.org/imc/2021/accepted/ https://conferences.sigcomm.org/imc/2021/accepted/ One of those papers is about counting the scale and scope of online political advertising during the 2020 election. How does one reproduce that study? The 2020 election is long past, and that data isn't archived anywhere other than what the researchers have already collected. This is a pretty simple empiracal data collection tastk, but you can't just re-measure that today because that study is about a moving target.
- tlb 5y ago
  I worked on this problem 25 years ago. It hasn't gone away. In general, performance comparisons are hard to reproduce. For instance, when benchmarking network protocols, often a tiny change in configuration makes a big change in a results. You might change the size of a buffer from 150 packets to 151 packets and see performance double (or halve.) Instead of making measurements with some arbitrary choices for parameters, you can take lots of measurements with parameters randomly varied to show a distribution of measurements. It's hard work to track down all the possible parameters and decide on a reasonable range for them, so it's rarely done. I found many 10x variances in network protocol performance (like how fairly competing TCP streams can sharing bandwidth). The big idea was to show that by randomizing some decisions in the protocol (like discarding packets with some probability as the buffer gets full) you can make the performance less sensitive to small changes. ie, more reproducible. Less sensitivity is especially good when you care about the worst-case performance rather than average. It can also make tuning a protocol much easier, since you aren't constantly being fooled by unstable performance. Performance sensitivity analysis is hard work, so most papers are just like "we ran our new thing 3 times and got similar numbers so there you go."
derbOac 5y ago
There are many causes of the lack of concern, but I think at the heart the problem, at least in the US, is that science has become politicized such that attempts at reform are mischaracterized for political gain. There's also a bit of ignorance in the general public, but that's only part of it. For example, if some on the right suggest some difficult but needed reforms, it tends to be spun as an attack on science. Or complaints that trivial projects are being overfunded get misinterpreted by the right and they try to make an example of the wrong studies for the wrong reasons. The pandemic was a good example of this in my mind, in that I think there were some serious systematic problems in academics and healthcare that were laid brutally bare, and many people suffered or died as a result. But then the whole thing got misidentified and sucked into the political vortex and all you end up with are hearings about how to rehabilitate the CDC, as if that is the problem and not a symptom of even bigger problems. I still think there are ways for things to change, but the most likely of them involve unnecessary suffering and chaos.
- N1H1L 5y ago
  I can give a different perspective. It is not because of politicization IMO - at least not in the hard sciences. The problem comes from way up, from Congress because the immediate impact of science is not obvious. Especially, for basic sciences the impact takes decades to be really felt. But then how do you do promotions? How do you judge output? Worse still, how does US Congress justify spending taxpayer dollars. Rather than acknowledging that any short term measurement of the quality of science is a fool's errand, we have doubled down on meaningless metrics like impact factors and h-indices. And this is what we have as a result.
thechao 5y ago
If you're any good at your chosen specialty you get a "feel" for the bullshit. I know this doesn't help the public. My experience is in medical research, crystallography, and computer science. Here's an example for detecting "bullshit" in cardiology: call up the MD PI from the published paper and ask to review anonymized charts from patients targeted with the procedure. Are there any? Then, the research is probably good; are there none? It's probably because it'd kill the patient. Similarly, in Programming Language Theory: we'd just ask which popular compilers added the pass. Is it on in -O3 in LLVM? Serious fucking result; is it in some dodgy branch in GHC? Not useful.
refurb 5y ago
It's not really broken because fraud gets discovered eventually. When I was working on my PhD, different lab were known to have different rigor. If a new discovery was made, skepticism was the norm. Until reproduced and proven, a new discovery was suspect. I mean it's clear that there is benefit to stamping out fraud just because it's inefficient, but in the end, the scientific method filters it out.

rossdavidh 5y ago

I think there's two problems impeding our ability to focus better on this: 1) for many people, the idea that science has widespread fraud is just hard to accept; in this respect it is similar to the difficulties that many religious communities have in accepting that their clergy could have a corruption problem 2) the solutions require thinking about problems like p-hacking, incentives, selection effects, and other non-trivial concepts that are tough for the average person to wrap their heads around.

SubiculumCode 5y ago
Also: I have not seen evidence of widespread fraud. Evidence of fraud,yes. Evidence of widespread fraud no.
- mistermann 5y ago
  Don't forget though: events proceed evidence, and evidence doesn't always follow events. Also: perception is ~effectively reality.
- derbOac 5y ago
  That's a fair point, although fraud per se is only a small part of all the problems. There's other forms of corruption than fraud, and a lot of it falls into this zone of plausible deniability rather than outright fraud. Also, I think the problems tend to find most weight with higher concentration of power, so what matters isn't as much "how widespread is corruption?" but rather "how is corruption distributed among power structures in academics and what is rewarded?"
- rossdavidh 5y ago
  Agreed it's an important point that fraud is only a fraction of the problem.
- bhk 5y ago
  I have. According to [1], "1 in 4 cancer research papers contains faked data". As the article argues, the standards are perhaps unreasonably strict, but even by more favorable criteria, 1/8 of the papers contained faked data. Interestingly, [2] using the same approach, found fraud in 12.4% of papers in the International Journal of Oncology. More broadly, [2] found fraud in about 4% of the papers studied (782 of 20,621). I'd say that's pretty widespread, but you further have to consider that these papers focused narrowly on a very specific type of fraud that is easy to detect (image duplication), so we would expect the true number of fraudulent papers to be much higher. [1] https://arstechnica.com/science/2015/06/study-claims-1-in-4-cancer-research-papers-contains-faked-data/ https://arstechnica.com/science/2015/06/study-claims-1-in-4-... [2] https://www.biorxiv.org/content/biorxiv/early/2016/04/20/049452.full.pdf https://www.biorxiv.org/content/biorxiv/early/2016/04/20/049...
derbOac 5y ago
I've often thought religious corruption is a good analogy, in that many of the societal dynamics are very similar. As I'm writing this the parallels are interesting to think about relative to US politics.
- throwawayboise 5y ago
  It it is a good analogy. For most lay people, science is a religion. They lack the expertise to understand the theory, but they unquestioningly accept the explanations and interpretations of the so-called experts. Most people don't understand astronomy and physics well enough to prove to themselves that the earth orbits the sun and not vice-versa. Yet they believe it does, with certainty, because they have been taught that it is true.

ArnoVW 5y ago

Aside from reproducibility issues in ML, what sort of issue did you have in mind in CS? Most CS work is 90% maths, I don't see how you can have reproducibility issues?

the_snooze 5y ago
Take, for example, network measurement research: https://conferences.sigcomm.org/imc/2021/accepted/ https://conferences.sigcomm.org/imc/2021/accepted/ One of those papers is about counting the scale and scope of online political advertising during the 2020 election. How does one reproduce that study? The 2020 election is long past, and that data isn't archived anywhere other than what the researchers have already collected. This is a pretty simple empiracal data collection tastk, but you can't just re-measure that today because that study is about a moving target.
tlb 5y ago
I worked on this problem 25 years ago. It hasn't gone away. In general, performance comparisons are hard to reproduce. For instance, when benchmarking network protocols, often a tiny change in configuration makes a big change in a results. You might change the size of a buffer from 150 packets to 151 packets and see performance double (or halve.) Instead of making measurements with some arbitrary choices for parameters, you can take lots of measurements with parameters randomly varied to show a distribution of measurements. It's hard work to track down all the possible parameters and decide on a reasonable range for them, so it's rarely done. I found many 10x variances in network protocol performance (like how fairly competing TCP streams can sharing bandwidth). The big idea was to show that by randomizing some decisions in the protocol (like discarding packets with some probability as the buffer gets full) you can make the performance less sensitive to small changes. ie, more reproducible. Less sensitivity is especially good when you care about the worst-case performance rather than average. It can also make tuning a protocol much easier, since you aren't constantly being fooled by unstable performance. Performance sensitivity analysis is hard work, so most papers are just like "we ran our new thing 3 times and got similar numbers so there you go."

derbOac 5y ago

There are many causes of the lack of concern, but I think at the heart the problem, at least in the US, is that science has become politicized such that attempts at reform are mischaracterized for political gain. There's also a bit of ignorance in the general public, but that's only part of it. For example, if some on the right suggest some difficult but needed reforms, it tends to be spun as an attack on science. Or complaints that trivial projects are being overfunded get misinterpreted by the right and they try to make an example of the wrong studies for the wrong reasons. The pandemic was a good example of this in my mind, in that I think there were some serious systematic problems in academics and healthcare that were laid brutally bare, and many people suffered or died as a result. But then the whole thing got misidentified and sucked into the political vortex and all you end up with are hearings about how to rehabilitate the CDC, as if that is the problem and not a symptom of even bigger problems. I still think there are ways for things to change, but the most likely of them involve unnecessary suffering and chaos.

N1H1L 5y ago
I can give a different perspective. It is not because of politicization IMO - at least not in the hard sciences. The problem comes from way up, from Congress because the immediate impact of science is not obvious. Especially, for basic sciences the impact takes decades to be really felt. But then how do you do promotions? How do you judge output? Worse still, how does US Congress justify spending taxpayer dollars. Rather than acknowledging that any short term measurement of the quality of science is a fool's errand, we have doubled down on meaningless metrics like impact factors and h-indices. And this is what we have as a result.

thechao 5y ago

If you're any good at your chosen specialty you get a "feel" for the bullshit. I know this doesn't help the public. My experience is in medical research, crystallography, and computer science. Here's an example for detecting "bullshit" in cardiology: call up the MD PI from the published paper and ask to review anonymized charts from patients targeted with the procedure. Are there any? Then, the research is probably good; are there none? It's probably because it'd kill the patient. Similarly, in Programming Language Theory: we'd just ask which popular compilers added the pass. Is it on in -O3 in LLVM? Serious fucking result; is it in some dodgy branch in GHC? Not useful.

refurb 5y ago

It's not really broken because fraud gets discovered eventually. When I was working on my PhD, different lab were known to have different rigor. If a new discovery was made, skepticism was the norm. Until reproduced and proven, a new discovery was suspect. I mean it's clear that there is benefit to stamping out fraud just because it's inefficient, but in the end, the scientific method filters it out.