In the first part of this post I mentioned some of the "bad" things during PhD studies, to summarize:
1) Grad students are "abused" by professors (not by all, but a good number of them nevertheless) and students are kept from graduating a little (or not so little) longer.
2) The assistanship (money you get to pay for expenses during grad school) is very little.
3) Physics is hard, even if you like it, you are in for stressful times.
There are probably more issues that affect a grad student, but those are the more universal ones I've seen. It doesn't end there though, once you obtain your PhD there are many more issues to deal with, a very important one is the job you will perform.
Many physics students (with no solid data, I would say that at least 70% of them) enter grad school with the hope of ending up in academia or something similar doing research that they like. Some also like to teach but mainly at a more advanced level than high school math or physics.
Life's a bitch, so the actual percentage of PhDs will not end up doing that. You will end up working as an engineer (if you are experimental) or as an advanced data analyst. A good number of my friends ended up working for either a finance firm or an oil-related company. It's true the money is good, and I think that's the reason they don't complain as much, but they're not doing what they wanted to do.
ISP proposes making a 2-year Master's degree in Physics more popular. After all, those who choose this path will not spend many years in school and might not regret working on something that's not physics. In a way I have always thought that a Master's degree in physics is useless so I see a few problems with this option. Here's why:
First, if I decide to go for a Master's degree I am basically throwing away any chance (as low as it can be with a PhD) at teaching at an advanced level or at research, and if I really want to do that then a Master's is not the way.
Second, assuming that you have thought about the first point and decided to give up that small opportunity at doing research by ending your student career at a Master's degree, why would you get a Master's in Physics? The only reason I can think is because you really have no idea what to do with your life. This is probably more common than people think, at least among physicists. So if you are "lost", you can go for a Master's degree which will (hopefully) give you a couple more years to mature and to find out what you want to do. But, what if you know what you want to do? You must remember that at this point you have given up on teaching or researching at college/university level. Well, if you know what you want to do, I would suggest you go for a Master's degree in that field. If you want to do finance, get at least an MBA, if it's engineering then go to an engineering college. You will learn what's relevant to that field and you will probably get a bigger check just because of your specialized background.
In my opinion no one should just go for a Physics Master's degree (OK, maybe science school teachers, but that is it). If you start the PhD and then you decide you don't want that for you anymore then a Master's degree in Physics is not a consolation prize nor any less than a master's in a different field and you should get your degree and leave. Plus taking the degree will at least help fill the two years otherwise missing on your resumé. But to just enter grad school to pursue a MS in Physics doesn't seem like the best idea to me.
Now, back to the topic of a PhD in Physics being the right choice. It's OK to get a PhD in Physics, it's fine if the students are willing to take the risk of giving up years of their lives for a small chance at a dream job (read academic job). What's not OK is that a lot of those students are not told what's waiting for them (the lab I am in right now has 3 grad students, none of us knew anything about what to expect in and after grad school).
Sure, Harvard and other top schools don't have to say much. After all, it's people from those schools that most likely will get the faculty job. But most of the PhD granting universities are not top, and they still need the grad students to teach labs, grade HWs and to some extent do the research (professors are too busy writing proposals). Those departments will flat-out lie to you promising a bright future. I know, I've been in one of those places.
If after being told the truth about your future you still want to go for a PhD, then go ahead, you will probably have fun and won't be disappointed. However, I get the feeling many of you will read this post a bit too late...
3 days ago