Jorge Hirsch...what an idiot. He came up with the h-index....the index which rates a scientist based on the number of articles they have with h number of citations. He claims this is a measure of achievement. Thats just retarded. First of all, in his article which establishes this h-index he says that he will avoid the discussion of the optimal method to rate "achievement", instead saying that he only wishes to show that h-index can be used as a predictor of future...wait for it..."achievement". Hes already establishing that citations = achievement and while it may be true in some cases...its dead wrong in others.
If I wanted to get something cited thousands of times...its quite simple. I just need to develop a computational software package...show that it works better or just as well as some other package for some need...and distribute it for free. In all honesty, the package that I develop will probably have already existed in some form or another and I simply did a systematic study of it to show it's relevance and yet...it will be cited an incredible amount of times simply because people will use it out of convenience.
...how about the papers which were the precursors to the development of GFP. Barely cited at all...not because they werent groundbreaking in some way...but because the paper at the very end of the chain which actually established the uses and benefits of GFP was so much higher impact that all of that...and yet...if you were to take the h-index of this...it would be very low. (this won the nobel prize).
The reasons I still feel that the JIF factor is better is simply because...journals like these accept papers which truly hold impact...some kind of weight...relevent to the problems that people are working on...in the moment. Because in the end, only others who are working on similar problems can recognize what is truly beneficial and novel...compared to the information which is nice to have..but not groundbreaking. And while this system has flaws itself...its much better than turning the scientific community into a game in which you try to get the most points...because honestly..there are methods to get papers cited more than others...and spending a life developing software for others to use doesnt make your "achievements" any greater than the person who used your software to develop a drug delivery system specific only to one situation (so it isn't cited much) but useful in saving the lives of thousands of people. and high impact journals can make the distinction between these two scenarios...not the h-index
Saturday, May 23, 2009
Subscribe to:
Post Comments (Atom)
4 comments:
Firstly, I'd like to point out that while the citation method won't include all good achievements, it reflects the obvious desire in our society to make everything uniform. We don't like to consider outliers. We tend to aggregate everything as much as possible to "simplify" experiences into something "understandable"(which usually doesn't actually achieve what it's meant to achieve in the first place.)Funny that I use the word achieve there (haha)
Second, I must defend his method somewhat in the sense that there is definitely some credibility there. Most of the highest achievers in society tend to be those that produce the most things, if only purely for the fact that the more you output, the more likely you are to produce a good piece of work. That isn't to say that there aren't people who tend to produce tons of bad things, or people who only produce one good thing, but there have actually been many studies done on the matter, which show that for a LARGE LARGE percentage of contributors to any field, this is the case. Once again, that does exclude a few though, in efforts to make data more "understandable."
In any case, I lost my train of thought, but back to citations, obviously you know already that the point in that method is to judge relevancy. theoretically the more a paper is cited, the more relevant it is to a field because other people will use it for their own research.
anyway, too sleepy. goodnight
theory may not be positively correlated with reality. it may not be negative correlated. it may not be correlated at all.
The reason I still hate his method despite all of your points which I can easily accept is simple. I think that his method would largely turn fields of study into a game. If you aren't sure if a somewhat controversial study will be cited...why pursue it when you could just take a safe methodical route that you know will interest x amount of people.
It just think it will encourage exactly the opposite of what they are trying to encourage right now in the scientific/engineering world which is a need for creativity and innovation in solutions. I just think the whole system biases things towards one type of thinking (and I generally think making everyone do things the same way is boring =P)
But who knows...maybe even if employers begin to pay attention to it, people may not change their attitude towards their work. What about all the underground citing rings that might pop up though!!! "Hey, I'll cite you if you cite me" kinda stuff =P
And then what do you end up doing with those methodology papers that get cited 1000s of times before someone points out that they're wrong and then the activity buzzing around them gets as cold as ice. Are those still good? Do you still consider those?
I dont know...I'm sleepy as well...I just came back from a huge bunch of research presentations (I had to go first) and the professors there just wouldnt let me go to the point where they just ended up having a discussion among themselves and I was just awkwardly standing up there...
even if the theory was once popular, and then later proved(or at least thought to be) incorrect, it still provided some contribution to the field, if only as a building block.
of course there's bound to be bias, and it's true that our type of thinking tends to be quite "in the box" so to speak.
so, after saying all that, what would you do to change it?
Post a Comment