I’m not saying anything new, but I’ll add to the chorus anyway.
A STEM BS degree at the sorts of universities you cite is primarily based on full-time coursework, and a good fraction of that coursework is “in major”, though with a good number of breadth-ensuring requirements in parallel.
A STEM PhD degree is primarily based on research. Some limited coursework is common but not universal. These courses may be required or elected to round out some areas of knowledge, say. The number of courses during a PhD could be zero or could be ten. It’s not going to be a ton, and it’s not going to be full-time coursework like it is during the BS degree. The degree might take five or six years, and after the second year it would be unusual to be taking any courses.
For physics (and certainly some other fields), an MS degree ahead of time is not typical but it does happen. Students who got their BS at a top US university essentially never get an MS before applying to physics PhD programs. Students with insufficient background or from different systems (e.g., a 3-yr program in some other country) or heading into a very technical area might benefit from a couple of years of additional coursework to make them better prepared (and thus competitive) at top schools. Thus, some students do enter PhD programs after finishing an MS program. At my program, I’d estimate maybe 15% or so.
I referenced GPT knowing it would rattle your cage, Stranger. lol Actually, I enjoy consulting GPT at times simply to see how absurd its answer might be. It scores a 100 in providing authoritative-sounding answers (at least, to the uninitiated), but scores poorly in accuracy or analytical rigor.
Here is my experience, with two graduate programs for CS for me (I left the first when my advisor died, but had jumped through all the hoops,) my wife’s in Biology and my daughter’s in psychology.
Hardly anyone entered with a Masters. (I did in the second program since I got one from my first school before I left, but it was not common.) You often spin off a masters on the way, and you usually get one if you are going to leave for some reason.
The departments I knew had quals based on knowledge of the field. You took classes on the subject of the quals if your undergrad program didn’t cover it. (MIT did not stress numerical analysis, so I took that class.) Once you passed those, you typically only took classes that interested you or more likely seminar classes reading papers on particular subjects. I was invited to sit in on one at Stanford for years, where the famous professors students gave reports on their research and I asked nasty questions. Hardly difficult. And of course there was a class for doing your dissertation. I wonder if the reason for that was to get you adequate hours to still be a student under the rules of the university.
No expectation of taking classes outside your field, unless you wanted to, I suppose.
MIT while I was there required undergrads to take humanities classes, but I doubt that extended to grad students. If you don’t know western culture by the time you get a BS it is too late.
Basically that. Student loan deferrals, international student visa eligibility, and other administrative issues can require full-time student status, and the easiest way to implement that for PhD students is to call research a “class”.
One other complication is that a graduate student pursuing a PhD will have a lot of “credit hours” on their transcript that are called something like “Doctoral Reading and Research”, which aren’t what you’d usually think of as “courses”, but represent the research and writing that you’re doing for your thesis. Those may or may not count for your purposes.
Unfortunately, many people glom onto the first part of your observation and ignore the second one, because getting “authoritative-sounding answers” is the goal even if the result is hot garbage.
I was previously of the opionion that while “AI” isn’t going to be capable of replacing “knowledge workers” (i.e. people who know stuff and produce critical analysis) it could at least serve as a conversational interface to knowledge-based information systems, but I am increasingly of the opinion that it isn’t even reliable enough for that application and will end up mangling basic factual information rather than just retrieving and parroting it. But people are so convinced that these systems are ‘good enough’ or have some complex world model in them which with just the right tuning can give accurate answers that they will implement them in critical roles and applications, and then not understand why medical treatment plans and court filings are full of superficially-plausible gibberish.
It’s been a ong time, but my recollection that there was not much in the way of course requirements for a PhD. I did take several courses to prepare for the prelims but once that was out of the way, it was mainly research. Also, in those bygone days there were two language reading exams, but I got those out of the way ASAP.
Just in case anyone isn’t sure: You don’t pay tuition for a PhD in STEM, and in fact are paid a stipend. There will be a tuition remission paid to the university on your behalf out of somebody’s coffers (grant, fellowship, department) but not out of yours.
I also want to point out that sometimes there are required courses that are major-adjacent as well. For example, I had to take industrial and electrical engineering courses, as well as a whole boatload of math (~17 hours IIRC) in order to fulfill my degree requirements, even though none of those were specifically Computer Science department courses, and nor were they University core curriculum courses either. There was also a technical writing course in there as well, and a minor field of study of 18 hours (they’ve since shrunk it for a while, and eventually eliminated it entirely)
So probably out of 135 hours, I’d guess that about 70-75 were required by the department.
There ought to be grants and what-not for just about any grad degree. I got a MBA and a MS at a different large state university, and I had a TA-ship, a grant, and a scholarship of some kind.
Between this sort of stuff and my experience with my Tesla Full Self Drive, I am confident our AI overlords have a long way to go before they replace us.
For the OP - never had to worry about it myself, but I vaguely recall that a course for PhD level was the same as a course for MSc? The difference was the degree (sorry) of difficulty in any thesis produced.
Especially given that virtually all universities have their course catalogs online and those tell you exactly what the degree requirements are, there’s no point in resorting to asking an AI chatbot.
Right, a graduate course is a graduate course. There are no courses (in the usual sense) that a PhD student would take that a masters’ student wouldn’t.
A graduate student might also teach some courses at a university, especially if they’re in a field that doesn’t have a lot of grant money.
It could mean they are a teacher, where in many districts you can get a $10k bump in salary for getting a masters. My sister and brother in law both have masters in Biology and Physics respectively for exactly that reason. Both got masters part time from programs that were chock full of identically situated people. The quality of their masters degree institutions was very, very low indeed. Think of getting your BS from UGA and then your masters from the online program at Valdosta State (I don’t know if such a thing exists). Accredited, but not very demanding. They literally learned nothing new.
For my PhD I had No required courses. I had to take 6 graduate seminars but the particular topics were not specified (4 in my department, 2 outside) which were basically directed discussions with the professor and about 6 other grad students around a particular topic. My committee required me to take a couple courses based on my thesis topic and my background. But most of my PhD was doing research not coursework. This was in the USA.
I am now a professor with appointments in 2 different departments across 2 different colleges, and also in several graduate centers, and coursework requirements for PhD students vary wildly across programs. Even though they’re all in the same University. There are no absolutes.
Note - I did have to pass an oral comprehensive examination at the end of my first year. They assumed that I didn’t need coursework to to do this, that I was a grad student and could teach myself anything I was deficient in. They were correct.
Also, never trust ChatGPT to tell you truth about ANYTHING.
This is what I remember from when I was in school.
A BS in biochemistry required about 45 hours of chemistry coursework, plus about 12 hours of biology coursework. 10 hours of physics, 8 hours of calculus. The rest was humanities to get to 120 hours.
A BA in biochemistry was only about 25 hours of chemistry coursework. I think math and physics were the same, it was just 20 hours less chemistry and 20 hours more of the humanities.
In other words, you get truthiness rather than truth.
To be precise, you don’t get a stipend for being a PhD student, per se, but you are awarded a teaching assistantship or research assistantship which comes with a stipend and tuition waiver. At least that’s how it worked when I was in grad school.
This is still true for all quality schools and PhDs in my experience (several of the next generation in my family are going through this now). If you aren’t offered an assistantship when you are admitted to a PhD program, run away.
You can get doctorates that are not research based, most commonly in Education and Fine Arts, where teaching professions pay premiums for Masters and Doctorates and may be de facto for high school Principal jobs and school superintendent of assistant superintendent positions. These are not funded by the degree granting institution are often part time, funded by either the student or their employer. These are typically not styled PhDs.
However there are many non PhD doctorates that are research based and variably funded, and many more that are “professional” degrees, most prominently medical and allied fields (MD, DDS, DO, DVM, etc) which are not funded by fellowships of assistantships.
That really depends on the college and the degree. Grants and scholarships are available for grad students just as they are available for undergrads - but you wouldn’t say that the existence of them for undergraduates means that no one should pay tuition. In certain fields/degrees , you can say that a student shouldn’t be paying tuition and if you are, it’s a sign that you shouldn’t be there at all but that’s only for certain degrees/fields. Most students in an MBA, Psy.D or MSW program are not fully funded and the same goes for teachers getting a Master’s for a pay bump. Some in those programs get grants and some have tuition paid by an employer but plenty do not.
That’s if you have a research or teaching assistantship. In my case my employer was paying for my degree, and they did indeed pay the standard per credit hour rate for everything including the 30 hours of research/dissertation credits.
ETA: I did have RA and TA jobs when I was getting my master’s as a full-time student, but (1) you were limited in term of classes per semester and (2) the pay was substistance level. I did get to have a faculty parking permit, though.