What level of knowledge should a new PhD graduate have?

[TomServo]

Let's say I'm getting a PhD in an area of gravitational physics. Upon getting that diploma, what is expected of me in terms of knowledge of GR? I hope my question can be useful to other students in other concentrations, so substitute QFT, or condensed matter, or whatever.

For example, is it expected that I am at least familiar with every topic in the Sean Carroll book? Or Wald? Or MTW? Is it bad if my quantum or classical e&m skills and knowledge have atrophied since I last took a course in them? Or is it just expected that I know my little area of gravitational physics expertly but it's okay if I need refreshers on other topics? I'm asking for somebody who is a newly minted PhD student, so what people expect from a postdoc applicant, for example.

I bought the book on relativity problems and I was going to try to solve them all this Summer. Is that too much misguided effort? I went to the April APS meeting and feel like I know AT BEST 5-10% of what everybody else know and that's just about gravity.

I hope my question doesn't sound stupid, thanks.

 

[romsofia]

It depends on what you did your research on. Gravitational physics is a huge fields. Was your thesis on numerical solutions in GR? Was it on the pursuit of quantum gravity? Did you explore non-general relativity approaches to gravity? This is important when trying to assess what you should "know".

No, you shouldn't be familiar on every topic in MTW, or Wald. Sean Carroll I would expect as that is a book for undergraduates. It is expected that you know the basics of QM, E+M, statistical mechanics and classical mechanics, and your field very well.

Buying a book to solve known solutions to problems in relativity isn't going to help, go read some research papers and discuss them with your group. If anything, I'd recommend this book if you're not sure with your skillset (https://www.amazon.com/dp/0521537800/?tag=pfamazon01-20).

Honestly, I don't think someone at this stage needs academic advice, but more so career advice. If a newly minted PhD didn't have a clue about what they know, and what they don't know, and where to go now that they're done, I'd be concerned about what they've been doing the last few years.

 

[TomServo]

It depends on what you did your research on. Gravitational physics is a huge fields. Was your thesis on numerical solutions in GR? Was it on the pursuit of quantum gravity? Did you explore non-general relativity approaches to gravity? This is important when trying to assess what you should "know".

Non-GR approaches to relativity, written a few papers being considered for publication so far.

No, you shouldn't be familiar on every topic in MTW, or Wald. Sean Carroll I would expect as that is a book for undergraduates. It is expected that you know the basics of QM, E+M, statistical mechanics and classical mechanics, and your field very well.

My problem is that I forget things if I go a long time without using them. I can usually get up to speed quickly with some study though.

Buying a book to solve known solutions to problems in relativity isn't going to help, go read some research papers and discuss them with your group. If anything, I'd recommend this book if you're not sure with your skillset (https://www.amazon.com/dp/0521537800/?tag=pfamazon01-20).

Thanks, I will get that book. This book:

https://www.amazon.com/dp/0691177783/?tag=pfamazon01-20

...was highly recommended by Thomas Baumgarte, a numerical relativity leader. I learn by doing lots of problems, and that helps ideas stick. Why do you think it won't help?

Honestly, I don't think someone at this stage needs academic advice, but more so career advice. If a newly minted PhD didn't have a clue about what they know, and what they don't know, and where to go now that they're done, I'd be concerned about what they've been doing the last few years.

Well actually I feel like I have another few years before I'm ready but my advisor thinks I could graduate by next Summer. My problem is I don't know if I have impostor syndrome or I'm genuinely not at the right stage of development. My advisor and others say I'm doing very well, should I assume they are correct?

 

[TomServo]

I should say I also learn by reading papers and trying to replicate them or go beyond them, so I am doing that as well.

 

[romsofia]

Non-GR approaches to relativity, written a few papers being considered for publication so far.

That is good, so the fact that you're able to produce papers that should get published is a good sign.

My problem is that I forget things if I go a long time without using them. I can usually get up to speed quickly with some study though.

There is an old saying, "If you don't use it, you lose it". This applies to knowledge as well, if you don't actively use a theorem, there is no reason you'd know how to use it immediately. You should always remember the concepts though, that's the important part. One thing I do is make my own notes out of courses/textbooks, so I know how I put it into my own words. If I look at a page in a book, or read back my notes from a lecture, and they look almost identical to the page or what my professor said, then I realize I didn't actually think about the concept, I just wrote it down and fooled myself into pretending I got what they were saying.

Thanks, I will get that book. This book:

https://www.amazon.com/dp/0691177783/?tag=pfamazon01-20

...was highly recommended by Thomas Baumgarte, a numerical relativity leader. I learn by doing lots of problems, and that helps ideas stick. Why do you think it won't help?

The way I approach research problems, and school work problems are very different. I've not checked out this book, but it seems to have good reviews from people I respect (Wheeler), so it can't hurt.

In my opinion, being able to derive the formulas used in physics is way more important than the minute calculations in a specific problem since we're at the research level, but maybe this book has a mixture of both. Like I said, it can't hurt, and we just approach learning differently.

Well actually I feel like I have another few years before I'm ready but my advisor thinks I could graduate by next Summer. My problem is I don't know if I have impostor syndrome or I'm genuinely not at the right stage of development. My advisor and others say I'm doing very well, should I assume they are correct?

Yes, I think you have imposer syndrome. I have it too, so just take the steps you need to re-affirm that you understand what you're doing. Graduate school is full of geniuses, and sometimes it looks like they just KNOW how to approach a problem instantly, meanwhile it would take me a few days on what to do. Just remember everyone has different backgrounds, so for all you know, they've just seen more problems than you have!

I would talk to your advisor on what he thinks the next steps in your career are if he wants you to graduate by next summer because you do seem a bit lost, which is understandable! Just trust your advisor/group and yourself, you should be fine.

 

[analogdesign]

Typically you do a Ph.D. on a narrow question in a wider field. If you're doing it right, when you graduate you should be the world expert (or close to it) on your exceedingly narrow question. After all, you spent years thinking deeply about it. You should also have a deep knowledge of your subfield (and have probably read hundreds of papers on it). You should also have a strong cross-cutting vision of your field in general and the place of your narrow question in the larger context. And, finally, you should have a thorough command of your overall field at the undergraduate level.

If I interview new PhDs and they aren't at this level I try to find out way. it isn't a good sign. Sometimes their advisor didn't help them find a narrow enough question, and sometimes they just aren't that strong and a kind-hearted committee signed off on their thesis since they put the years in.

All that said, you usually make a lateral jump to a new subfield or perhaps just a different narrow question in your first job. The PhD is your chance to learn how to learn (in a fundamental way) and your result can demonstrate that.

 

[CrysPhys]

And, finally, you should have a thorough command of your overall field at the undergraduate level.

<<Emphasis added>>

I guess a lot depends what you mean by "thorough" and "overall". But if we are talking about core physics topics (e.g., mechanics, E&M, QM, thermo & stat mech, along with requisite math), I would expect a PhD to have a strong command at the grad level.

 

[TomServo]

<<Emphasis added>>

I guess a lot depends what you mean by "thorough" and "overall". But if we are talking about core physics topics (e.g., mechanics, E&M, QM, thermo & stat mech, along with requisite math), I would expect a PhD to have a strong command at the grad level.

Crap, I’m rusty on topics I haven’t seen in two years. How does one stay sharp in areas unrelated to their concentration throughout their PhD?

 

[Vanadium 50]

How does one stay sharp in areas unrelated to their concentration throughout their PhD?

Colloquia

 

[symbolipoint]

Crap, I’m rusty on topics I haven’t seen in two years. How does one stay sharp in areas unrelated to their concentration throughout their PhD?

Colloquia

Makes me wonder. Review and repeated experience in whatever way one could get them should be of value. "Staying sharp" was given as the goal. Maybe new coursework, getting back to laboratory experiences by internship or job or also with taking upperlevel or grad level course, or something which compares.