Strange Difficulties with Advisor

[manofpeace]

TLDR: I'm somewhat worried that my primary advisor will write a LOR that significantly underrepresents my work, behavior, and character.

I am currently a master's student in statistics. My research advisor has recently been advising me avidly that it is a bad idea to apply to PhD program in statistics with the end goal of industry. They claim that everyone who applies to PhD's in statistics have the end goal of academia, and if they end up in industry it's because something went wrong. They have stated that my interest in industry, my planned gap year working in industry post-graduation, and my uncertainty about whether I am going to apply and are reasons why I should not apply for a PhD, and further that I would not get into a good one if I did, since it would somehow be evident from my application.

This seems somewhat wrong based off of the other professors and students that I've talked to. I wanted to get opinions from a broader audience, and would much appreciate any thoughts on the following:

• I'm worried that they may not write me a good letter of recommendation. I've proved a theorem related to machine learning (I think-we remain to find an error) and I have always been polite and respectful. However, I naively mentioned another one of their advisees who is also very bright and serious who wants to go into industry after a PhD and they said, "Maybe he won't get in anywhere!" I truly think they have good intentions, but I'm still worried, as they are the only professor I've done research for as a grad student. Further, they have not gotten around to checking my proof so I'm not sure they really know what I've done. I've asked cordially if they could check it and they just replied that I should know how to check my own work (even though I've checked it ~15 times). They seemed excited about this theorem before I came up with a proof, but now they do not and have even talked down on it. They have also been saying that taking hard math classes are more important than research, which is the opposite of what I have heard generally.

Thanks in advance

Edited October 7, 2022 by manofpeace

[boxyroxy]

It is indeed distasteful for a professor to speak to you that way. However I would heed your professors advice about advanced math classes if your goal is a phd. Seems like you need to start treating this as a business transaction. Your professor seems to think phds all end up in academia unless they wind up in industry by mistake. You want a letter from him, then you should tell him you want to go into academia (whether you actually do, is up to you - 5 years after you start your phd).

If your professor truly believes people end up in industry due to something going wrong, then your professor must believe things go wrong for at least half the statistics graduates in the US. His thinking is very backwards and really not grounded in reality. I would be concerned for the students being advised by this professor.

[bob loblaw]

With all due respect, your advisor seems like the archetype of an out-of-touch academic (probably a boomer). Sorry to hear this.

 If I were you, I would not let him hold you hostage, ignore sunk-costs and start establishing relationships with professors you've taken courses from. It's not the end of the world to have mediocre LORs.  I personally did not have great LORs but it was fine for me. 

Like the previous comment said, however, having a firm mathematical foundation is important (especially in a program that emphasizes probability). Also if you're uncertain about a PhD program BEFORE you apply, those thoughts are gonna be amplified once you're in the program. I'd keep that in mind. 

[manofpeace]

This is helpful, I appreciate your responses.

I take it from your posts that applying for PhD's in statistics with the end goal of industry research is actually a valid/common goal?

My other question is if I'm applying to PhD's in biostatistics instead, is it still as important to take as many advanced math classes as possible?

Edited October 10, 2022 by manofpeace

[bob loblaw]

@manofpeace 

Absolutely. Having a PhD is obviously a prerequisite now for any research oriented industry role.  But also given the increased competition for data science roles, I think a PhD will be valuable there as well.  

WRT to math, it's not all or nothing.  My view is that you should at least have taken proof-based linear algebra and real analysis at a minimum. Having other classes like stochastic processes, convex optimization, measure theory, etc. will make you a more competitive candidate.  Of course, the math classes you choose to take should make sense given your research interests.  For example, taking an abstract algebra course may kind of seem random for some but may make sense if your interest lie in random matrices.

Different programs (bio stats included) care about mathematical background to varying degrees: more theoretical a given department is (which tends to be higher ranked or whatever), the more they'll care.  It also depends on how competitive your application year happens to be: UCLA Biostats, for example, weeds people out by mathematical background especially in competitive years.  My program is more on the applied side so it doesn't care that much. 

Edited October 10, 2022 by bob loblaw
More explanation and typo

[bayessays]

22 hours ago, manofpeace said:

I take it from your posts that applying for PhD's in statistics with the end goal of industry research is actually a valid/common goal?

Not only is it common, but it's MUCH more common than academia (and not just as a backup).

[BL4CKxP3NGU1N]

On 10/10/2022 at 10:29 AM, manofpeace said:

I take it from your posts that applying for PhD's in statistics with the end goal of industry research is actually a valid/common goal?

One big reason I chose the program that I am currently attending is because the majority of PhD grads from my department go into industry and have great placements. My goal has pretty much been industry from the beginning, and the same is true for probably about half of my cohort. 

On 10/10/2022 at 10:29 AM, manofpeace said:

My other question is if I'm applying to PhD's in biostatistics instead, is it still as important to take as many advanced math classes as possible?

Biostatistics PhD programs tend to be more applied than Statistics programs, so as long as you have the basics (like calc 1-3, linear algebra, real analysis), you shouldn't worry too much about taking a lot of additional advanced math classes just to have them on your transcript. A lot of this is program dependent though, so just make sure you understand the level of mathematical rigor at each program you're looking at applying to by reading through the list of required courses. Keep in mind though that having a deeper math background can go a long way in putting yourself ahead of other applicants.

[manofpeace]

It's possible I may be misunderstanding him - what he might mean is that PhD students at very top schools such as Stanford, Berkeley, etc. all/mostly enter their programs aiming towards academia. Maybe this would this make more sense?

Edited October 15, 2022 by manofpeace

[Counterfactual]

3 hours ago, manofpeace said:

It's possible I may be misunderstanding him - what he might mean is that PhD students at very top schools such as Stanford, Berkeley, etc. all/mostly enter their programs aiming towards academia. Maybe this would this make more sense?

The observation is based on small sample size, but I do know a Stanford and several Berkeley PhD students who decided or are thinking about going into industry upon graduation. So I guess it's hard to say.

[manofpeace]

20 hours ago, Counterfactual said:

The observation is based on small sample size, but I do know a Stanford and several Berkeley PhD students who decided or are thinking about going into industry upon graduation. So I guess it's hard to say.

Interesting, are they relatively early on in their programs or closer to graduating?

[Counterfactual]

Most of them are in their third or fourth year when we met. Though in terms of career path, to the best of my knowledge, at least two were determined to go into industry before matriculation.  (But I could imagine that they probably did not explicitly show that on their SOP/PS)

So yeah, I'd say there's still time to figure that out along the way during PhD if one's not sure. But if you already know that you wanted to go into industry, there are people in top programs that have the same goals too.

Even so, from my humble experience, it never hurts to have an open mind career-wise.

Edited October 16, 2022 by Counterfactual