Masters to top PhD program?

[CyberShot]

I am curious to know if someone from a physics Masters program at a state school (high GPA, excellent research) has the the same chance of admission to Harvard, Princeton, or MIT PhD program vs someone with similar stats from undergraduate. How hard is it to get into top 10 schools from Master's programs? Can the Masters program serve as a kind of clean slate for someone who messed up his/her undergrad physics GPA, and wants to do well in the Masters program to get into a top PhD program?



P.S. Please no responses like "it's unlikely you will succeed in the Master's program after you bombed your undergrad."


Thanks.

 

[Anonymous217]

I'm curious too. Not necessarily about the "M.A. compensating for poor undergrad performance", but the comparison between applicants graduating with a M.A. and applicants graduating with a B.A.

I'm planning on graduating early to get my Bachelor's, but I figure I could just solidify my chances to top grad schools by staying the "extra" year to start and finish a Master's.

 

[Pengwuino]

If they really are the same kind of student that is hard working, same GPA, same research experience, same types of letter of reccs and (P)GRE scores, they stand a pretty similar chance of getting in. School ranking doesn't matter much in physics.

 

[CyberShot]

If they really are the same kind of student that is hard working, same GPA, same research experience, same types of letter of reccs and (P)GRE scores, they stand a pretty similar chance of getting in. School ranking doesn't matter much in physics.

What do you mean school ranking doesn't matter much?

I would think that a 3.8 GPA from MIT > 3.5 from UC Santa Cruz


Also, do you think someone with a 4.0 MA GPA from a California state school, good publishing record, and an excellent PGRE has a good chance of being admitted to a top 5 physics grad program?

 

[Pengwuino]

What do you mean school ranking doesn't matter much?

I would think that a 3.8 GPA from MIT > 3.5 from UC Santa Cruz

Uhm... those aren't similar GPAs. I don't see what you're trying to say

Also, do you think someone with a 4.0 MA GPA from a California state school, good publishing record, and an excellent PGRE has a good chance of being admitted to a top 5 physics grad program?

Yes, except there is no such thing as "top 5 physics grad program". Physics graduate schools aren't ranked meaningfully so it's a meaningless statement.

 

[twofish-quant]

I would think that a 3.8 GPA from MIT > 3.5 from UC Santa Cruz

I wouldn't say this. I don't have any experience with UC Santa Cruz, but I do have experience with both MIT and UT Austin, and I'd have to say that UT Austin grades its undergraduate physics students more harshly than MIT does.

 

[twofish-quant]

I am curious to know if someone from a physics Masters program at a state school (high GPA, excellent research) has the the same chance of admission to Harvard, Princeton, or MIT PhD program vs someone with similar stats from undergraduate

I'll defer to someone with more specific information if they contradict this, but it's extremely rare for graduate students to transfer physics programs, so if you are trying to transfer with a masters degree, you are going to find yourself at a disadvantage to start with.

How hard is it to get into top 10 schools from Master's programs?

I don't know what a "top 10 physics school" is.

Can the Masters program serve as a kind of clean slate for someone who messed up his/her undergrad physics GPA, and wants to do well in the Masters program to get into a top PhD program?

No. Also I don't see what's with this desire to get into a "top Ph.D. program." It's not as if you should expect a job as a professor going out or anything like that.

P.S. Please no responses like "it's unlikely you will succeed in the Master's program after you bombed your undergrad."

With a few exceptions, most physics degree programs are Ph.D. programs with the masters degree being a consolation prize.

 

[twofish-quant]

I'm planning on graduating early to get my Bachelor's, but I figure I could just solidify my chances to top grad schools by staying the "extra" year to start and finish a Master's.

In every physics program that I know about, in the unlikely event that you get admitted with a masters degree you are going to have to basically retake everything from scratch.

 

[Pengwuino]

In every physics program that I know about, in the unlikely event that you get admitted with a masters degree you are going to have to basically retake everything from scratch.

I have to wonder about this statement. I got accepted into 2 phd programs with an MS.

 

[twofish-quant]

I have to wonder about this statement. I got accepted into 2 phd programs with an MS.

You must have gotten in at one that I don't know about then... :-) :-) :-)

 

[CyberShot]

I'll defer to someone with more specific information if they contradict this, but it's extremely rare for graduate students to transfer physics programs, so if you are trying to transfer with a masters degree, you are going to find yourself at a disadvantage to start with.

I'm not sure why you call it "transferring." Isn't the masters degree separate from the PhD? You graduate with a masters, and the PhD is a whole different degree, and thus you need get re-matriculated. (Unless you're starting a PhD to begin with)

I don't know what a "top 10 physics school" is.



No. Also I don't see what's with this desire to get into a "top Ph.D. program." It's not as if you should expect a job as a professor going out or anything like that.

I meant according to US News Ranking. And let's be honest, even though some may question its validity, getting your PhD from Princeton, Harvard, or MIT is much more prestigious than from UC Santa Cruz, or UofArizona (even if it is in radio astronomy).

The reason to get a PhD from a top 10 school is that, statistically speaking, you have a better chance of getting hired as faculty, and the prestige opens more doors in terms of postdocs (working with more established faculty, higher chance of working with renowned advisor, better equipment, more funding, etc).

If two people have very similar stats when applying for a postdoc, but one of them got a PhD from a top tier school, you'd be inclined to higher that student. It's not really fair, but it's really the best metric the hiring committee has, (a lot like the SAT).

There is a reason those top schools are hard to get into. I don't really understand when people say not to look at prestige when it comes to graduate school, because in the end, it really does matter.




P.S. As for my above post, I meant to say that

"3.8 UC Santa Cruz < 3.5 MIT" GPA-wise

 

[Anonymous217]

In every physics program that I know about, in the unlikely event that you get admitted with a masters degree you are going to have to basically retake everything from scratch.

This may be true. But graduating early with a B.A. and getting a M.A. is certainly preferable over staying the extra year just for the B.A. Certainly if the M.A. course credit does not transfer over, then doing so for the extra year in the B.A. wouldn't either. Oh, and I'm pure math by the way. I'm sure most credit would be able to transfer over though. And if not, I wouldn't mind taking a couple courses the first two years of the Ph.D.

 

[Choppy]

I'm not sure why you call it "transferring." Isn't the masters degree separate from the PhD? You graduate with a masters, and the PhD is a whole different degree, and thus you need get re-matriculated. (Unless you're starting a PhD to begin with)

It really depends on where you are. In Canada, we more often start with a master's program and either finish or transfer into a PhD program. In the US, I believe it's far more common to start in a PhD program. In the UK and European systems it often seems that a master's degree is more of an extension of undergraduate work. (This is a VERY broad generalization though.)

I meant according to US News Ranking. And let's be honest, even though some may question its validity, getting your PhD from Princeton, Harvard, or MIT is much more prestigious than from UC Santa Cruz, or UofArizona (even if it is in radio astronomy).

The reason to get a PhD from a top 10 school is that, statistically speaking, you have a better chance of getting hired as faculty, and the prestige opens more doors in terms of postdocs (working with more established faculty, higher chance of working with renowned advisor, better equipment, more funding, etc).

If two people have very similar stats when applying for a postdoc, but one of them got a PhD from a top tier school, you'd be inclined to higher that student. It's not really fair, but it's really the best metric the hiring committee has, (a lot like the SAT).

There is a reason those top schools are hard to get into. I don't really understand when people say not to look at prestige when it comes to graduate school, because in the end, it really does matter.

For someone who hasn't even gone to graduate school you seem to know a lot about how the academic system works.

I could be wrong on this, but part of the criteria for some of these rankings is the number of graduates a program produces. Thus, if you survey the institutions that current faculty graduated from and find a higher proportion are from these 'top ranked' programs, the data is skewed from the fact that they've produced more graduates.

You also have to account for individual quality of students and academic bottleneck. The 'top tier' schools tend to have more applicants and thus you may introduce a sampling bias from the pool of students coming in. Higher quality in, usually means higher quality out - a factor which is independent of the actual program quality.

One bias I've seen in hiring post-docs is that graduates from a school's own program tend to be favoured. In the hiring committees I've been on, school prestige has not played any more of a role than that. Where candidates are more-or-less equal 'stat-wise' the favoured candidate is chosen based on factors such as how well the interview goes and how well the candidate appears to fit in with the group.

P.S. As for my above post, I meant to say that

"3.8 UC Santa Cruz < 3.5 MIT" GPA-wise

Can you point to a conversion scale for us? I highly doubt this.

 

[deRham]

The reason to get a PhD from a top 10 school is that, statistically speaking, you have a better chance of getting hired as faculty, and the prestige opens more doors in terms of postdocs (working with more established faculty, higher chance of working with renowned advisor, better equipment, more funding, etc).

I used to think like this, but I am inclined to believe it is a muddled and confused attitude. I believe the causation goes precisely the opposite way - people who tend to have the firepower to become established faculty often (but not always) get accepted to top schools.

As for renowned advisers, I also used to think the best advisers are at the best US News schools. How incredibly wrong I was. It seems like there is truth to US News in terms of the top schools when you define 'top' to include a lot of schools, WAY more than 10.

The bottom line is that if you're incredibly productive a researcher, and work on things people in academia care about, you'll get somewhere. If not, I don't think distinguishing yourself with a top 5 school name matters in the least.

Those who get into the absolute hardest to get into schools are often doing so many things right that it does work out pretty well for them.

But many times, at the bottom of the heap of say, top 5 schools, someone will have gotten in over his or her friend only because of some special perk that only mattered then and there, and won't really afford an advantage in the long run.

 

[deRham]

If two people have very similar stats when applying for a postdoc, but one of them got a PhD from a top tier school, you'd be inclined to higher that student.

No, probably you'd go with the one whose research strikes your fancy and with regards to your university's needs - do you need more people in X or Y field. Unless both of them did something so generic as to end up seeming extremely similar.

Now if your letters of reference come from people of vastly different reputation, then maybe the one of reputation would carry more weight.

I think it highly unlikely that it would come down to the school's name itself, although name in other senses does matter. More likely, if someone did go to a highly lowly ranked school, it's quite possible s/he might just not have produced something quite as impressive.





As justification for what I am saying, by the way (and I agree I could be slanting slightly too much in favor of disregarding the top US News program's name, but I don't think much) - I have found a trend when I examine faculty at various schools, whereby below the so-called top of the top schools, you no longer find the professors to have largely homogeneous background.

It's true that many of the faculty at so-called tier 1 schools come from tier 1 undergrad backgrounds, but they probably were not only tier 1 school students, but actually top of the top even among their peers. Some of them were probably in the league of the best mathematicians in their field in their given country. Below that, I think little things such as school name stop mattering very quickly, because almost anyone in the running went to a pretty darn good school already, simply because they were always showing promise. But at that point, factors such as whose work is in demand at a specific place, luck, etc all start playing a lot more role than what school they came from. It's very, very conceivable in the long term professorship game, to me, for someone coming from a lower ranked US News school to get picked over someone from a higher ranked one, simply because the odds of success for BOTH are so low, and the expectation of the competition's ability is so high that it stops depending on their school of origin.

 

[deRham]

Can the Masters program serve as a kind of clean slate for someone who messed up his/her undergrad physics GPA, and wants to do well in the Masters program to get into a top PhD program?

Yes and no. It depends if the evidence you show of success explains that what you did before is no longer representative of your ability or drive.

 

[twofish-quant]

I'm not sure why you call it "transferring." Isn't the masters degree separate from the PhD? You graduate with a masters, and the PhD is a whole different degree, and thus you need get re-matriculated. (Unless you're starting a PhD to begin with)

That's not how it works for US physics/math/astronomy programs. The standard path is that you get admitted with a bachelors degree into a Ph.D. program. Once you finish the course work, they hand you a masters.

UK/Canada apparently is different, and it's also very different in other fields. My wife has a Ph.D. in education and in that field it's standard for people to get a masters, go to work, and then get a Ph.D. some years later at a different school.

I meant according to US News Ranking. And let's be honest, even though some may question its validity, getting your PhD from Princeton, Harvard, or MIT is much more prestigious than from UC Santa Cruz, or UofArizona (even if it is in radio astronomy).

U Arizona is optical. It's U Virgina that is radio astronomy...

Among people that don't have a clue about physics, maybe, but among physicists (i.e. people that will hire and give you jobs), it's not.

First of all, no one really cares about what school that you went to. They care a lot about who your dissertation adviser was.

Second, the "big name schools" in physics are often not the "big name schools".

Sometimes there are physics reasons for this.

For example, the reason that University of Arizona and University of Hawaii are just much better schools for optical astronomy than MIT is because it makes more sense to put a telescope in the middle of the desert or on a high mountain in Hawaii than it does to put one in downtown Cambridge. If you go to Arizona or Hawaii, you will be able to use the telescopes they have without fighting (as much) for time.

The other thing is that certain people are at certain universities. If you are doing late time stellar evolution models and you go to UC Santa Cruz, I'm going to be impressed, since it means that you are learning from one of the gods of the field.

The reason to get a PhD from a top 10 school is that, statistically speaking, you have a better chance of getting hired as faculty, and the prestige opens more doors in terms of postdocs (working with more established faculty, higher chance of working with renowned advisor, better equipment, more funding, etc).

Can you show me these statistics?

Something that may be misleading is that if you look where faculty got their degrees from, you'll see a lot of big name schools, but if you look at the dates, you'll see that a lot of those degrees were from the 1960's and 1970's. In the 1960's, the big name schools were the only game in town so to speak, but what ended up happening is that the faculty from those schools ended up starting departments in the mid-West.

One consequence of this is that you end up connected to the networks where ever you end up.

If two people have very similar stats when applying for a postdoc, but one of them got a PhD from a top tier school, you'd be inclined to higher that student. It's not really fair, but it's really the best metric the hiring committee has, (a lot like the SAT).

Maybe, but "top tier school" in physics is different from "top tier school" according to USNWR. If I knew nothing about a Ph.D. other than they went to SUNY Stony Brook or Harvard and the post-doc was for nuclear physics, the SUNY Stony Brook person would get hired because in that area it's the "bigger name."

Also, since you are probably not going to get a research professorship anyway, I don't think that increasing your odds should matter much.

There is a reason those top schools are hard to get into.

And curiously it's harder to get into Hawaii than it is to get into MIT.

I don't really understand when people say not to look at prestige when it comes to graduate school, because in the end, it really does matter.

And you know this how?

Prestige *does* matter, but USNWR is a horrible measure of prestige when it comes to physics graduate schools. If you want to know which physics schools are good, then go into the academic literature, surf the web, and if you find a bunch of papers and professors that excite you, then they will likely excite other physics people.

Also since you've mentioned UC Santa Cruz, you should know that one of the "gods of supernova nucleosynthesis calculations and late time stellar evolution codes" happens to work there.

Also you can change the prestige. The thing about most graduate schools is that they are tiny. If there is a graduating class of two, and you are one of them, you can radically change the prestige of the university that you graduate from. Once you get to Ph.D. programs, you change the prestige of the university more than the university changes you.

"3.8 UC Santa Cruz < 3.5 MIT" GPA-wise

Again, I have no idea about UC Santa Cruz, but in my experience, MIT grades more harshly than UT Austin, so someone that gets a 3.5 at UT Austin would in my view be "sharper" than someone that gets a 3.8 at MIT. If you want me to compare UT Austin with MIT GPA's, I'd give a small correction factor.

Now one reason I don't consider GPA that important is that if you give me a UC Santa Cruz student, I have no clue what the correction factor is.

UT Austin is an example of how stats can be misleading. Personally, I think that the median UT Austin student is less mathematically sharp than the median MIT student *however* my experience has been that the media UT Austin student in science and engineering is just as competent as the median MIT student in science and engineering.

One thing that I'd like to know from you is "who is telling you this?" You are saying things that make no sense to me, and I'd like to find out who is giving you this bad information so that I can have a talk with them.

 

[CyberShot]

That's not how it works for US physics/math/astronomy programs. The standard path is that you get admitted with a bachelors degree into a Ph.D. program. Once you finish the course work, they hand you a masters.

UK/Canada apparently is different, and it's also very different in other fields. My wife has a Ph.D. in education and in that field it's standard for people to get a masters, go to work, and then get a Ph.D. some years later at a different school

That may be the standard path, but not everyone follows the standard path, right? Isn't it the applicant's choice whether to apply to the PhD or Masters program? If that's the case, then I'd recommend less than stellar applicants to "redeem" themselves in the Masters program, graduate with an MS. And then apply specifically to the PhD portion of some program.

U Arizona is optical. It's U Virgina that is radio astronomy...

Among people that don't have a clue about physics, maybe, but among physicists (i.e. people that will hire and give you jobs), it's not.

First of all, no one really cares about what school that you went to. They care a lot about who your dissertation adviser was.

Aren't most of the big name advisors to be found at top schools, anyways?

For example, the reason that University of Arizona and University of Hawaii are just much better schools for optical astronomy than MIT is because it makes more sense to put a telescope in the middle of the desert or on a high mountain in Hawaii than it does to put one in downtown Cambridge. If you go to Arizona or Hawaii, you will be able to use the telescopes they have without fighting (as much) for time.

The other thing is that certain people are at certain universities. If you are doing late time stellar evolution models and you go to UC Santa Cruz, I'm going to be impressed, since it means that you are learning from one of the gods of the field.

Well, the geography of it makes sense. But I get the feeling that top schools like HYPS are better suited to study more important, fundamental topics like string theory, or HEP because those are the questions that humanity cares more about (and should rightly so), especially because these schools have a history of being funded by the government during times like WW2, and the Cold War. Optical astronomy and the sort is limited only to our solar system and its neighbors. Why should we limit our capabilities to study one infinitesimally small subset of the entire Universe? What merits does that have? Sure, it may give us small clues in the right direction, but ultimately we should be addressing issues such as the Big Bang, cosmic inflation, where did we come from?

The minutia should follow naturally from that.

Can you show me these statistics?

Don't have any at the moment, per se, but I feel confident that a lot of professors would agree that it definitely matters where you got your degree from. I would still think that most nuclear physicists would prefer to get their PhD from MIT even though Michigan State University is ranked #1 in nuclear. In my limited experience, physics departments are somewhat superficial in that they would like to have faculty with PhD's from 'prestigious' places, otherwise their own reputation would suffer.

I also read from a much older thread on this forum that if you go to a top tier school your chances of tenureship are about 1 in 4, and if you go to a top 25~50 school then it's somewhere like 1 in 20. But I'll try to find that thread.

And curiously it's harder to get into Hawaii than it is to get into MIT.

?

Also you can change the prestige. The thing about most graduate schools is that they are tiny. If there is a graduating class of two, and you are one of them, you can radically change the prestige of the university that you graduate from. Once you get to Ph.D. programs, you change the prestige of the university more than the university changes you.

Yes, and most of the people who've changed the prestige of their universities are the ones who've won Nobel Prizes (who are in turn the ones who attended top tier schools) Now, it's likely that they had the capabilities to succeed all along and the school had nothing to do with it, but people get excited about attending a school where you can brag about your advisor's very low Erdos number

One thing that I'd like to know from you is "who is telling you this?" You are saying things that make no sense to me, and I'd like to find out who is giving you this bad information so that I can have a talk with them.

No one specifically. I'd just like to think I'm representing the general population of the forum that would probably not think twice about accepting a top tier offer.

 

[Pengwuino]

That may be the standard path, but not everyone follows the standard path, right? Isn't it the applicant's choice whether to apply to the PhD or Masters program? If that's the case, then I'd recommend less than stellar applicants to "redeem" themselves in the Masters program, graduate with an MS. And then apply specifically to the PhD portion of some program.

I followed this path and I think it got me into a better PhD school than it would have if I had applied directly for a Phd after my BS. However, it really was because I practically "found religion" my very last semester as an undergrad and began to work hard into my MS while having an abyssmal undergrad record.

Aren't most of the big name advisors to be found at top schools, anyways?

No. Think about this for a second. Departments only have so much money to pay people. A very recent and well known example around where I live was Raymond Chiao. UC Merced, a new school with no reputation at all, flat out stole him from UC Berkeley by offering him enough money and lab space. Rice University recently got 3 top UCSD biochemist or biomedical physicists to leave by offering them something like $500k a year and brand new lab space.

No one specifically. I'd just like to think I'm representing the general population of the forum that would probably not think twice about accepting a top tier offer.

If Caltech or Stanford offered me a position at their PhD programs, I would throw it in the garbage (actually I'd frame it). Neither school has anything I want to do. In fact, besides Princeton and UT Austin, not many high prestige universities work in the field I want to work in. Remember, you're not getting your PhD in physics as much as you're getting your phd in a sub-field of physics, despite what your diploma tells you :)

 

[twofish-quant]

Isn't it the applicant's choice whether to apply to the PhD or Masters program?

Depends on the department. For UT Austin astronomy, they will not admit people for a terminal masters, and that is common. It's also impossible as far as I can tell to request admission "only" to the Ph.D. part of the program.

Aren't most of the big name advisors to be found at top schools, anyways?

By top meaning USNWR rankings, no.

But I get the feeling that top schools like HYPS are better suited to study more important, fundamental topics like string theory, or HEP because those are the questions that humanity cares more about (and should rightly so), especially because these schools have a history of being funded by the government during times like WW2, and the Cold War. Optical astronomy and the sort is limited only to our solar system and its neighbors.

Why should funding in 1950 matter today? Funding matters a lot, but there is a lot of political pressure to make sure that science money gets spread evenly across the country.

Also, a department can get stuck in a rut. One criticism of Harvard and Princeton is that their departments are too heavy into string theory, and so there is no room for alternatives like loop quantum gravity. And the center for LQG research seems to be Louisiana State.

Why should we limit our capabilities to study one infinitesimally small subset of the entire Universe? What merits does that have? Sure, it may give us small clues in the right direction, but ultimately we should be addressing issues such as the Big Bang, cosmic inflation, where did we come from?

You answer the big questions by look at the small details. Also it's really tough to do theory if you don't have any real data.

Don't have any at the moment, per se, but I feel confident that a lot of professors would agree that it definitely matters where you got your degree from.

We have professors here in this forum. Let's see what they think.

Also, I don't think no one here thinks that it doesn't matter where you do your Ph.D. research. The question is whether USNWR means anything, and since the people that put together USNWR aren't physicists and have no clue about physics, I don't think that what they think really matters.

I would still think that most nuclear physicists would prefer to get their PhD from MIT even though Michigan State University is ranked #1 in nuclear.

We have nuclear physicists in this forum. Let's ask them.

I can talk about computational astrophysics reputations. Harvard, MIT, and Princeton have very good reputations in this area. Yale and Stanford don't have any particular strength in this area.

In my limited experience, physics departments are somewhat superficial in that they would like to have faculty with PhD's from 'prestigious' places, otherwise their own reputation would suffer.

That's not been my experience. It is true that graduates from well-connected schools have an advantage because they have stronger social networks, but I don't see how raw prestige really factors much.

Also one thing that curiously works *against* prestige in astronomy is the fact that so many people have Harvard and Princeton connections. If you look at astronomy departments, a lot of them were started in the 1970's by Harvard and Princeton graduates, so Harvard and Princeton are in some ways the "mother ship." However, what ends up happening is that because you have so many departments run by Harvard and Princeton grads, it's not hard for a Ph.D. to get "connected" into that social network.

Also because you have so many people with connections trying to build up their departments, people will react badly if people from the "mother ship" use their connections excessively.

This is only for astrophysics theory in the United States. Other fields and other countries might be very different. I know that MBA's work in a very different way, as does physics schools in China, where there is a very strong tier system.

Yes, and most of the people who've changed the prestige of their universities are the ones who've won Nobel Prizes (who are in turn the ones who attended top tier schools)

No there aren't. It's not that hard to become a "big name" in your field. For example in supernova research there are maybe 50 or so people in the world that does this, and everyone knows everyone else. If you are a post-doc that publishes good papers, you will get noticed. Maybe no one outside of those 50 people will know or care who you are, but so what?

For example, in supernova research, Florida Atlantic University happens to be a big mover and shaker. There is a brilliant professor there, who had a brilliant student that is now a major "mover and shaker" in the world of supernova research. Same for UC Santa Cruz.

It's also not that hard for a small university to become a center of excellence in some field.

Now, it's likely that they had the capabilities to succeed all along and the school had nothing to do with it, but people get excited about attending a school where you can brag about your advisor's very low Erdos number

People with no particular experience in physics.

Also ultimately, it's about you and not your adviser. One thing about being a good student is that ultimately if you are a good student that increases the prestige of your adviser.

No one specifically. I'd just like to think I'm representing the general population of the forum that would probably not think twice about accepting a top tier offer.

No is saying that you shouldn't apply to Harvard. No one is saying that you shouldn't take the offer if they give it to you. The trouble is that you can apply, you probably won't get in.

Now what?

The other problem is that you will be in for a lot of trouble if you don't know what really is top tier. If you want to study LQG, skip Harvard and go for Louisiana State.

 

[twofish-quant]

I followed this path and I think it got me into a better PhD school than it would have if I had applied directly for a Phd after my BS. However, it really was because I practically "found religion" my very last semester as an undergrad and began to work hard into my MS while having an abyssmal undergrad record.

Also this is a different situation. If you have a horrible undergraduate record, then having a masters degree in which you end up being decent will help a lot. It won't erase the past, but it will help a lot. The other thing is that in general, coursework in masters won't transfer.

The other thing is that there is a difference between terminal masters and non-terminal ones.

Remember, you're not getting your PhD in physics as much as you're getting your phd in a sub-field of physics, despite what your diploma tells you :)

And in astrophysics, prestige tends to follow the adviser. If you have a big name adviser go to North Podunk University, then people will say cool NPU must be great because so-and-so works there.

One other way of thinking about it is that you are getting a Ph.D. in your own tiny subfield of physics in which you are the #1 world expert on the topic of whatever your dissertation is.

One criterion for getting the Ph.D. is that you get the Ph.D. when you can convince your committee that you know more about the topic of your dissertation than they do.

 

[deRham]

Aren't most of the big name advisors to be found at top schools, anyways?

How do you define "top" ? I agree USNews does a fair job of capturing where MOST of the big name advisers will be if you widen your gaze from, say, top 10 to top 25 or 50. There will still be many amazing people who don't get caught in this range, but at least in my field, you can be nearly certain that at top 25, someone is doing what you want, and doing it well.

You'll actually be surprised - the profiles of the people at the so-called top 10 aren't necessarily 100% better than those at the rest. However, the concentration of absolute geniuses is highest at those.

I have seen plenty of cases of absolutely brilliant mathematicians from foreign universities becoming professors at various U.S. schools - not necessarily the top 5. And I think one of the above posts explains a likely reason - probably, they were just given a better offer by a certain university.

 

[twofish-quant]

I agree USNews does a fair job of capturing where MOST of the big name advisers will be if you widen your gaze from, say, top 10 to top 25 or 50.

Also "big name" doesn't necessarily mean that anyone outside of the field has heard of them. The way that physicists work reminds me a lot of gossiping grandmothers or high school cliques, and some people within the small group of gossips have higher reputations.

One reason that it's a good idea to familiarize with the physics literature in the field that you are interested in, is that you should be able to tell for yourself what is a good paper.

You'll actually be surprised - the profiles of the people at the so-called top 10 aren't necessarily 100% better than those at the rest. However, the concentration of absolute geniuses is highest at those.

Absolute mathematical genius is surprisingly unimportant in computational astrophysics.

And I think one of the above posts explains a likely reason - probably, they were just given a better offer by a certain university.

Senior scientists are a lot like chefs opening a restaurant (sometimes with egos to match). This makes offers by small no-name universities attractive.

If you are a "physics star" and Harvard offers you a position, then you still have to do things Harvard's way. If North Podunk University offers a "physics star" a faculty position, then those people can run their department however the hell they want.