2 ms·

A Flurry of Copycats on PubMed

31 points by beefman 12y ago

nn3 12y ago
Good example of Cambell's law. Doesn't seem much worse than the senior professor in western departments who automatically gets co-authorship on everything his department produces. Publication indexes are just very sick metrics. http://en.wikipedia.org/wiki/Perverse_incentive http://en.wikipedia.org/wiki/Perverse_incentive
jrapdx3 12y ago
Though I use PubMed frequently, can't say I've encountered "copycat" articles. It's disheartening to find out what has been happening, but perhaps not a huge surprise to the extent publication becomes merely a commodity to accumulate in order to further an academic career. It does prompt me to keep an eye out for such occurrence. But I'm pretty sure not every researcher goes about things that way. Maybe it depends on the field of research. The topics of interest to me could simply be too obscure to attract that kind of attention. Sure seems if anything can be done with computing technology, it will be done. What sort of future awaits us? We humans always seem to do until we overdo then redo it all over again, it doesn't ever end.
jldugger 12y ago
A friend of mine reached out to me via LinkedIn several months ago, stating he could probably use Tableau and some public datasets to DDoS the medical peer review system and earn tenure at the same time: low impact factor * a million is still going to be a lot better than the alternative. Sadly, I didn't hear anything terribly insightful when I asked him about spurious correlations and data mining as a perjorative term, so I left that one alone, and I assume he doesn't have the balls to go through with it since you know, there are actual people in the tenure committee who have to review your CV. One angle that isn't quite clear to me from the article; if you build a system to harvest datasets, find correlations, and push them through a set of templates with conditional subtemplates based on saliency, with correct citations of source data to produce a single article, is it wrong to put your name on it? And if you do 30 of them? Obviously selling your naming rights to papers harms both the community who might wish follow up with the authors on scholarship, and those who use peer reviewed publications as a filter. But I think the real lesson here might just be that abusing the university administration's institutional data analysis platform to game your tenure case seems like a waste of time in comparison to selling one weak article a day for 10,000 bucks.

31 points by beefman 12y ago

nn3 12y ago

Good example of Cambell's law. Doesn't seem much worse than the senior professor in western departments who automatically gets co-authorship on everything his department produces. Publication indexes are just very sick metrics. http://en.wikipedia.org/wiki/Perverse_incentive http://en.wikipedia.org/wiki/Perverse_incentive

jrapdx3 12y ago

Though I use PubMed frequently, can't say I've encountered "copycat" articles. It's disheartening to find out what has been happening, but perhaps not a huge surprise to the extent publication becomes merely a commodity to accumulate in order to further an academic career. It does prompt me to keep an eye out for such occurrence. But I'm pretty sure not every researcher goes about things that way. Maybe it depends on the field of research. The topics of interest to me could simply be too obscure to attract that kind of attention. Sure seems if anything can be done with computing technology, it will be done. What sort of future awaits us? We humans always seem to do until we overdo then redo it all over again, it doesn't ever end.

jldugger 12y ago

A friend of mine reached out to me via LinkedIn several months ago, stating he could probably use Tableau and some public datasets to DDoS the medical peer review system and earn tenure at the same time: low impact factor * a million is still going to be a lot better than the alternative. Sadly, I didn't hear anything terribly insightful when I asked him about spurious correlations and data mining as a perjorative term, so I left that one alone, and I assume he doesn't have the balls to go through with it since you know, there are actual people in the tenure committee who have to review your CV. One angle that isn't quite clear to me from the article; if you build a system to harvest datasets, find correlations, and push them through a set of templates with conditional subtemplates based on saliency, with correct citations of source data to produce a single article, is it wrong to put your name on it? And if you do 30 of them? Obviously selling your naming rights to papers harms both the community who might wish follow up with the authors on scholarship, and those who use peer reviewed publications as a filter. But I think the real lesson here might just be that abusing the university administration's institutional data analysis platform to game your tenure case seems like a waste of time in comparison to selling one weak article a day for 10,000 bucks.