Perhaps you have heard the expression, "If you want something done, ask a busy person." Some people are able to get a lot done, and adding one or four more things to the list doesn't slow them down. Let's call this type of person Type W.
Then there are Type X people, and I am going to classify them, for the sake of discussion and Pseudo-Scientific Rigor, as Type X1 and Type X2. Type X people -- in general, with all other factors being equal -- don't get as much done as Type W people.
To explain the difference between X1 and X2, and to compare them with W people, I will use an academic example. I am not thinking of any particular person or people; this is a hypothetical situation. I am not (necessarily) talking about you.
Imagine 3 graduate students at similar stages of their academic program, with similar types of research and similar backgrounds and the same advisor. They are done taking classes and can focus on their research.
A Type W person would get a lot done whether they were funded by a research assistantship (RA), a teaching assistantship (TA), a fellowship, or whatever.
A Type X1 person would only make decent research progress if funded by an RA or fellowship. A TA would consume all of X1's time and energy, not because X1 is more devoted to teaching than W, but because X1 can only focus on one thing at a time.
A Type X2 person would get more done if partially funded by an RA or fellowship and partially by something requiring a bit of structured work -- for example, perhaps teaching one lab or discussion section, or perhaps doing some grading or other work like that. If funded entirely by an RA or fellowship, X2 wouldn't be able to deal effectively with the lack of structure and would waste a lot of time, making very slow progress, even if the advisor set specific goals.
Actually, I can think of one real example of W vs. X, and I have written about this before. Back in days of yore, my own graduate advisor gave an RA to another student instead of to me, saying that I would get a lot done even if I were a TA, whereas the other student would only get work done if an RA.
At the time, I felt like I was being punished for being a Type W person and the other student was being rewarded for being an X. Now, as a grad advisor doling out limited funds within the limited time frame of a grant, I can understand it better. I also want to mention that the other student in question was, and still is, my friend, and that I did not blame him for being an X-person.
Explanatory note about grad funding in my field: Many students are funded by a mixture of types of support over time; some RA, some TA, some fellowship. Advisors make decisions from year to year about the type of support for each student. If we requested 12 months of support for a student on a grant for the entire duration of the grant, the budget would explode, leaving no money for the actual research, so my colleagues and I typically ask for partial support, and make up the rest with TA or other sources of research support, or the student gets a fellowship, etc.
With that in mind, my question now is how (and whether) to distinguish between the X1's and the X2's.
I guess 'whether' is the more first-order question: In cases involving making choices for a particular grant/project, would you -- the advisor -- take into account work habits like the W vs. X scenarios described above when making decisions about support?
And if you do, would you make a distinction, like I did, between X1 and X2? Would you distinguish them by trial-and-error, or is there some magic formula you can use to predict (barring routine and unpredictable research setbacks) how cost effective someone will be? I think not, and I have mostly given up trying to guess.
My current strategy, which is not obviously better than anything else I have ever tried, is to accumulate as much grad support as possible, give students the benefit of the doubt as much as possible, distribute grad support in the way that makes the most sense for research and human resource priorities (what needs to get done when and by whom?), and hope for the best.
Nevertheless, when making some decisions and when trying to understand how people work best, I think it is useful to think about W vs. X, and more vs. less structure, and to explore ways to stretch grant funding to the maximum extent possible to cover as many students as possible for as long as possible. That is the goal. Would you also like to see me pull a rabbit out of a hat? Too bad, I can't do that either, and not just because I have no training or authorization for the use of magical animal subjects in research.