4 ms·

Actually, i look at from the Game theory viewpoint. What metrics are the rewards in an academic career based on?? Publications. It's rather simple that the res

by aangjie 15y ago

Actually, i look at from the Game theory viewpoint. What metrics are the rewards in an academic career based on?? Publications. It's rather simple that the researchers will find ways(shortcuts) to maximize that. Ofcourse, the Impact Factor is a new evolving measure too. But I think it's only harder to hack, as it needs social hacking skills.(That too very subtly via written text).

amirmc 15y ago
> "It's rather simple that the researchers will find ways(shortcuts) to maximize that." I agree with the issue about how academics are measured but I find this leap to be grossly unfair (and from experience, completely untrue). There is an issue that academics may not fully understand statistical methods. I think this is likely due to the more sophisticated experiments being developed that necessitate more mathematical rigour during analysis. Suggesting that academics are simply trying to 'game' the system is ludicrous. Also, I wanted to point to this comment on the OP. It talks about how a psychologist would have dealt with the experiment described. http://www.guardian.co.uk/discussion/comment-permalink/12344150 http://www.guardian.co.uk/discussion/comment-permalink/12344...
- aangjie 15y ago
  > Suggesting that academics are simply trying to 'game' the system is ludicrous. I am trying to suggest that academics are deliberately trying to game the system. Only observing that if it were a simulation a rational agent would follow that specific strategy. Luckily, most people are not perfect rational agents and so science progresses faster. Sorry my comment came across that way. > It talks about how a psychologist would have dealt with the experiment described. http://www.guardian.co.uk/discussion/comment-permalink/12344.. http://www.guardian.co.uk/discussion/comment-permalink/12344.... I spent studying cog. sci in a psychology department for two years. I know it's true, but to be fair, the amount of work involved in the experiments is huge too.
  - amirmc 15y ago
    > "I am trying to suggest that academics are deliberately trying to game the system" Did you mean "I am not trying to suggest..."? Otherwise I'm confused. In general, I don't think it's helpful to think of things like this from a purely rational, game-theoretic perspective. As you mention, humans are not rational & there isn't perfect information. Also, it involves too many assumptions about what might drive people (eg it's not just about publication lists).
    - aangjie 15y ago
      OOps, Yeah i meant "I am not trying to suggest".. //drives of people Yeah, we agree on that.
_delirium 15y ago
Of course, the Impact Factor is a new evolving measure too. But I think it's only harder to hack, as it needs social hacking skills. On an institutional level, some journals already have slightly-hush-hush semi-official gaming policies in place. Since citations from a journal's articles to its own back issues count when computing the impact factor, a common practice is to "suggest" to authors several relevant citations from back-issues of the journal to insert. Suggesting relevant citations is a longstanding practice, of course, but some journals now devote time to systematically finding self-citations to the journal itself, and suggesting them to (almost) every author, as a way of upping its impact factor. Journal grey-hat SEO, perhaps.
- amirmc 15y ago
  Also some academics routinely cite their own group's work wherever possible to bump up the citation numbers.

by aangjie 15y ago

amirmc 15y ago

> "It's rather simple that the researchers will find ways(shortcuts) to maximize that." I agree with the issue about how academics are measured but I find this leap to be grossly unfair (and from experience, completely untrue). There is an issue that academics may not fully understand statistical methods. I think this is likely due to the more sophisticated experiments being developed that necessitate more mathematical rigour during analysis. Suggesting that academics are simply trying to 'game' the system is ludicrous. Also, I wanted to point to this comment on the OP. It talks about how a psychologist would have dealt with the experiment described. http://www.guardian.co.uk/discussion/comment-permalink/12344150 http://www.guardian.co.uk/discussion/comment-permalink/12344...

aangjie 15y ago
> Suggesting that academics are simply trying to 'game' the system is ludicrous. I am trying to suggest that academics are deliberately trying to game the system. Only observing that if it were a simulation a rational agent would follow that specific strategy. Luckily, most people are not perfect rational agents and so science progresses faster. Sorry my comment came across that way. > It talks about how a psychologist would have dealt with the experiment described. http://www.guardian.co.uk/discussion/comment-permalink/12344.. http://www.guardian.co.uk/discussion/comment-permalink/12344.... I spent studying cog. sci in a psychology department for two years. I know it's true, but to be fair, the amount of work involved in the experiments is huge too.
- amirmc 15y ago
  > "I am trying to suggest that academics are deliberately trying to game the system" Did you mean "I am not trying to suggest..."? Otherwise I'm confused. In general, I don't think it's helpful to think of things like this from a purely rational, game-theoretic perspective. As you mention, humans are not rational & there isn't perfect information. Also, it involves too many assumptions about what might drive people (eg it's not just about publication lists).
  - aangjie 15y ago
    OOps, Yeah i meant "I am not trying to suggest".. //drives of people Yeah, we agree on that.

_delirium 15y ago

Of course, the Impact Factor is a new evolving measure too. But I think it's only harder to hack, as it needs social hacking skills. On an institutional level, some journals already have slightly-hush-hush semi-official gaming policies in place. Since citations from a journal's articles to its own back issues count when computing the impact factor, a common practice is to "suggest" to authors several relevant citations from back-issues of the journal to insert. Suggesting relevant citations is a longstanding practice, of course, but some journals now devote time to systematically finding self-citations to the journal itself, and suggesting them to (almost) every author, as a way of upping its impact factor. Journal grey-hat SEO, perhaps.

amirmc 15y ago
Also some academics routinely cite their own group's work wherever possible to bump up the citation numbers.