[poincaraux]

[ QUOTE ]
It is not up to the administrators to push the students.

[/ QUOTE ]

That's not really true. Here's an easy example: all of the schools that admitted me said something like this in their letter: "we'll guarantee you $20K/year plus tuition and fees for 5 years, assuming you remain a student in good standing." Now, I went out of my way to find funding sources, but what if I hadn't? You can bet that our secretaries would have pressed me to apply for fellowships, etc. rather than have the department pony up $45K/yr.

More generally, things run better with money. Good administrators will do what they can to get as much extra money as possible flowing into their departments. And things run way better when the grad students have enough money to get by. Good administrators help to make sure this is the case.

[ QUOTE ]
it usually up to the student and the PI to write the grant.

[/ QUOTE ]

That's more true in some fields than others. I'm in biophysics, and every grant I've seen has been written either by the PI or by a Post Doc. Grad students might help out a bit, but they don't do bulk of the work. I'm pretty sure it's the same in bioinformatics, but I'm not positive.

My wife, on the other hand, is in Ecology and Evolutionary Biology. She writes a ton of grants herself. Never the big multimillion dollar ones, obviously, but a lot of grants for a couple of thousand.

[ QUOTE ]
That said most grad schools pay their students (and tuition is deferred).

[/ QUOTE ]

In our department, most students are paid directly by their advisors, who are expected to have grants to cover such things. The tuition isn't deferred either; the advisors pay it.