I can't speak for CS, where the coverage isn't as good, but pretty much all scientific PhDs work in labs that are NIH funded. In most cases, students (foreign

I can't speak for CS, where the coverage isn't as good, but pretty much all scientific PhDs work in labs that are NIH funded. In most cases, students (foreign or otherwise) end up doing a TA for 2-3 years and then get funded by their PI's research grant. For example, in my lab every grad students stipend and tuition fee came out of my advisor's grants. Then again in the sciences, if you want to do any meaningful research you have to do a PhD. I was one of the few who was lucky to avoid doing a postdoc (which in most cases is mandatory to try and get a faculty or industry research position). Note that I'd do that PhD 100 times out of 100. Was a ton of fun.