Should I Do a PhD? Informed Opinion & Relevant Considerations

[epenguin]

Here, aimed at Britain but probably fairly generally applicable. Informed opinion, relevant considerations.

"
Should I do a PhD? You asked Google – here's the answer
Rachel Hall"

http://click.mail.theguardian.com/?...425796543893606f2bbb1f5555ce87256944cb183d9c3

 

[jedishrfu]

Interesting and quite true. I knew as early as 1970's that getting a PhD would be problematic. I had one math prof who was quite a good teacher lose out on tenure because he didn't publish enough papers for the department. The department was transitioning from a teaching role to a research role at a liberal arts college to retain other young mathematicians and he was on the losing end. To me it was a real loss to the college.

For him it worked out as he got into another school and eventually became the math department head while maintaining his excellent teacher rating among students. I think he published a few papers since that time.

Many PhDs find work in industry doing high tech sorts of things and never get back to academia as the pay level is better and the stress level is reduced.

 

[StatGuy2000]

The Guardian article is clearly taking a harsh stance with regard to graduate studies. Specifically consider this quote: "In some industries, a PhD might even set you back, as business leaders often see them as driving a largely pointless three-year wedge between an undergraduate degree and an entry-level position." I would be curious to see which industries this article is referring to.

Personally, I think the answer to whether one should do a PhD is highly dependent on which specific field. In some fields (for example, my field of statistics) earning a PhD is often considered the equivalent of a Masters degree plus an additional 2-3 years of work experience (since the research is often considered as work experience), which can be a valuable head start in landing high-paying positions to start with. Also, in statistics a PhD also opens options within industry (mainly research-level opportunities) which may not be as readily available for someone with just a Masters (at least not without many more years of work experience).

The situation may well be different for other fields such as, say, physics.

 

[PAllen]

The Guardian article is clearly taking a harsh stance with regard to graduate studies. Specifically consider this quote: "In some industries, a PhD might even set you back, as business leaders often see them as driving a largely pointless three-year wedge between an undergraduate degree and an entry-level position." I would curious which industries this article is referring to.

Personally, I think the answer to whether one should do a PhD is highly dependent on which specific field. In some fields (for example, my field of statistics) earning a PhD is often considered the equivalent of a Masters degree plus an additional 2-3 years of work experience (since the research is often considered as work experience), which can be a valuable head start in landing high-paying positions to start with. Also, in statistics a PhD also opens options within industry (mainly research-level opportunities) which may not be as readily available for someone with just a Masters (at least not without many more years of work experience).

The situation may well be different for other fields such as, say, physics.

Note, the very next paragraph discusses earnings premium from getting a PhD, consistent with what you say. However, while they give a reference for the earnings premium, they give no reference for the purported harmful case. So perhaps they should have skipped that comment altogether. Much of the article seems perfectly reasonable to me.

 

[russ_watters]

The Guardian article is clearly taking a harsh stance with regard to graduate studies. Specifically consider this quote: "In some industries, a PhD might even set you back, as business leaders often see them as driving a largely pointless three-year wedge between an undergraduate degree and an entry-level position." I would be curious to see which industries this article is referring to.

Personally, I think the answer to whether one should do a PhD is highly dependent on which specific field. In some fields (for example, my field of statistics) earning a PhD is often considered the equivalent of a Masters degree plus an additional 2-3 years of work experience (since the research is often considered as work experience), which can be a valuable head start in landing high-paying positions to start with. Also, in statistics a PhD also opens options within industry (mainly research-level opportunities) which may not be as readily available for someone with just a Masters (at least not without many more years of work experience).

The situation may well be different for other fields such as, say, physics.

I think you answered your own question just fine, because it really is straightforward. To generalize: A PhD will help you get a job that requires or desires a PhD, and will likely hurt you for a job that does not.

Take my industry: HVAC/building systems. I don't think I've met a PhD mechanical or electrical engineer in the industry. They don't want to work here and we would probably not want them. It's not rocket science.

...But a PhD is likely helpful for some rocket science jobs.

In any case, if you have to ask google a question that starts; "Should I...", the answer is almost certainly no.

 

[russ_watters]

Hmm... I thought of a reason why my views might not apply here and then realized it was exactly why they do:

Leaving academia and entering industry is a small to giant step toward engineering. But [pure] research and engineering are fundamentally different goals, so it is not a trivial thing to try to serve prospective PhD candidates both as if they were equivalent.

Taking someone who wants the freedom to try to expand the boundaries of human knowledge (with total security and autonomy) and telling them they need to start engineering products for sale (at the whim of the economy and their company) and shortly thereafter managing people isn't necessarily an easy sell. I'd think it would be rather like telling a concert pianist they can always fall back on being a DJ.

Of course this issue is different for different majors and people, though. Medical research, for example, has a tight marriage between academia and industry.

 

[Vanadium 50]

we would probably not want them.

Would that still apply to someone without a PhD who did something else unrelated for 3-5 years? (e.g. is this pure opportunity cost?) For example, Peace Corps, military, or stay at home mom?

 

[George Jones]

or stay at home mom?

or stay at home dad!, e.g., I have a colleague whose husband is a stay at home dad.

 

[russ_watters]

Would that still apply to someone without a PhD who did something else unrelated for 3-5 years? (e.g. is this pure opportunity cost?) For example, Peace Corps, military, or stay at home mom?

It might, but less. Any of those people would be cheaper than a PhD.

 

[Vanadium 50]

or stay at home dad

My implied point was one of disparate impact.

Any of those people would be cheaper than a PhD.

Is this a requirement? It is in, for example, some public school districts, where there is a formula for pay involving years of service and hours past a BS or MS. Someone just starting as a school teacher costs (in one district that has their contract on the web) 50% more with a PhD than someone just starting out. (And 20% more than someone with a BS and ten years experience)

Or is this an assumption? Along with the assumption "they don't really want to work here". (Which is still an assumption, despite the abundant validation one can find on PF)

 

[Crass_Oscillator]

Isn't the degree far too broad for blanket opinions to apply?

I've been in a graduate computer science department and graduate engineering department for my PhD and students often started doing internships during the degree where they applied their research to an industrial problem before landing a job in academia or industry. In my particular case there are academic and industrial jobs that would be difficult to acquire without the unique PhD skills I am acquiring.

Physics students working on non-falsifiable noodle theory may have a different situation ahead of them, and English students writing yet another essay about Virginia Woolf are probably in even more dire straits.

 

[russ_watters]

Is this a requirement? It is in, for example, some public school districts, where there is a formula for pay involving years of service and hours past a BS or MS. Someone just starting as a school teacher costs (in one district that has their contract on the web) 50% more with a PhD than someone just starting out. (And 20% more than someone with a BS and ten years experience)

Or is this an assumption? Along with the assumption "they don't really want to work here". (Which is still an assumption, despite the abundant validation one can find on PF)

I'm not 100% sure of the thrust of this question, so I'll give the short answer and then expand: It's an assumption.

The longer answer starts with it not mattering what is true and what is assumed because what matters is the actions taken based on those beliefs. If a resume comes across my desk for a no-experience junior project engineering position and the person is a brand new PhD physicist, I'll [wonder why HR didn't screen it out], say "huh?" and then probably not bother responding to it. If it includes a nice cover letter, I may read it, but it had better contain good answers to these questions:
1. Why do you want this job, vs one in your field?
1a. Are you just going to quit if you find a job in your field in 4 months?
2. Are you willing to accept what this job typically pays?

Someone with a BS in physics or engineering physics is much more likely to get a call than a PhD in physics. Yes, there is such a thing as "overqualified".

Note, I work for a consulting engineering firm, not the government, so payscales don't really exist. We bill our clients what they are willing to pay for a "junior project engineer" and we can't have one making double what is typical. We couldn't compete for projects or be profitable with someone like that.

I also work in a lower-end, commodity engineering field, so this may matter more than if you are a physicist looking for a mechanical engineering job in, say, aerospace. There's still a small practical skills gap, but they may be more willing to overlook that and pay a higher salary for someone proven to be really bright. But that's just a guess.

 

[davidwinth]

In my field, I would say it's not really worth it at all if you want to teach but don't want to be a major researcher. It looks like it may end up being a waste of time and money for me to have gotten a PhD in mechanical engineering so I could get into teaching science at university level. The only places that I can get a job are at community colleges where they would have taken me with my master's degree.

I did learn a lot, though! So on a personal level I am still happy that I did it.

 

[EngWiPy]

In general, to get into the academia you need long and lean years of intensive research and publications. In the industry, I think it depends on the field, in the first place. For example, a PhD in CS is not like a PhD in history or literature. Then it depends on the job market. I am not sure about Britain, but in the US the job market is vibrant and dynamic, and can absorb PhDs. That being said, in my personal situation, and based on the outcome of it, it wasn't worth it.

 

[Grinkle]

In the semi-conductor industry a PhD can help if you intend to pursue a technical career path. I know a person who co-authored a textbook in the area in which he was working in my company at the time (also the area he studied in his graduate work). I think him having a PhD made him more publishable, I am confident he could have written a fine textbook with or without the PhD but maybe (I speculate) not had it published without the academic credentials he also had. Anyway, this definitely helped his influence within the company as being an industry acknowledged expert in his field and it did help him achieve career goals for himself. In his case, it wasn't just having PhD behind his name, it was all about how he leveraged it later on by keeping his academic contacts and network current and finding ways to make all that useful to the company he worked for.

@russ_watters makes a great point about over qualifying yourself for many engineering jobs, so be sure you are ok with that aspect if you do pursue a PhD.

 

[Crass_Oscillator]

In the semi-conductor industry a PhD can help if you intend to pursue a technical career path. I know a person who co-authored a textbook in the area in which he was working in my company at the time (also the area he studied in his graduate work). I think him having a PhD made him more publishable, I am confident he could have written a fine textbook with or without the PhD but maybe (I speculate) not had it published without the academic credentials he also had. Anyway, this definitely helped his influence within the company as being an industry acknowledged expert in his field and it did help him achieve career goals for himself. In his case, it wasn't just having PhD behind his name, it was all about how he leveraged it later on by keeping his academic contacts and network current and finding ways to make all that useful to the company he worked for.

@russ_watters makes a great point about over qualifying yourself for many engineering jobs, so be sure you are ok with that aspect if you do pursue a PhD.

A PhD is essential for a technical career path in certain cases in semiconductors. It's the difference between being a user of the TCAD programs to design a device and being the scientist who wrote the TCAD program and designed the algorithms/models.

Some precocious masters degree holders may learn the advanced physics they would need on their own, but that is a very unlikely possibility.

 

[Grinkle]

@Crass_Oscillator Yes, I agree. Si process development and certainly CAD algorithm development involving device physics are areas that often do require a PhD.

 

[Zap]

I really don't think high starting earnings is a good reason to pursue a PhD, because while you may start with a higher salary, you'll also be completely broke for the five or more years you spend getting the PhD, so I really don't know if it actually pays off all things considered. I can't imagine pursuing a PhD for financial gain. The most common reason I see for pursuing a PhD is for citizenship. That actually makes sense, because you can easily get a visa as a graduate student. Honestly, white Americans are a minority in graduate school, and if you ask the foreigners why they are getting a PhD, the answer is usually as simple as to come to America.

The Americans that are in graduate school are usually those who just like school and want to keep doing the school thing, or they wanted get involved with some research that interested them or a combination of the two. Some are there because they want to be a university professor, but in my own experience, I find that most are not there due to the pursuit of any specific career path. A minority are those like myself, who are there due to mishap, poor planning and poor advice.

 

[jedishrfu]

Once you've chosen a field to pursue then you have to decide what degree you need in order to succeed and that target may change as study. Many fields are stratified where the only way to get the best salary is to be a PhD level professional and that may open doors to management.

In emerging fields, BS or MS degrees will be fine but as the field matures then MS becomes the degree to have and finally PhD supercedes them all. Its been that case for me as I've moved through computer science over the years. Stratified fields means there are salary ceilings that can't be broken through without a better degree or by becoming a manager.

Also it depends on the company you work for. Some companies have a professional ceiling where the only way to get more salary is to become a manager. Others have dual tracks where management and professional salaries can overlap meaning a professional could make more than his/her manager.

When making these decisions its best to speak with someone whose been through it and knows the field well enough to advise you.

 

[Zap]

Physics students working on non-falsifiable noodle theory may have a different situation ahead of them ...

One of my professors proudly talked about what he actually called spaghetti theory of the quantum vacuum, or something like that. For some reason, he came up with a theory that the structure of the vacuum was like spaghetti. I had no idea what he was talking about, but it's hilarious to see it mentioned here.

 

[EngWiPy]

In my personal experience, a PhD is waste of time and money if it's for financial reasons. Maybe a PhD in few fields can help, like in machine learning these days, but for most fields and jobs it's not beneficial at best. Sometimes it works the opposite direction: you can become unemployed, as with a PhD you cannot get junior positions to get experience, and for senior positions you don't have enough experience. It's a dilemma once you get into that situation.

 

[Zap]

In my personal experience, a PhD is waste of time and money if it's for financial reasons. Maybe a PhD in few fields can help, like in machine learning these days, but for most fields and jobs it's not beneficial at best. Sometimes it works the opposite direction: you can become unemployed, as with a PhD you cannot get junior positions to get experience, and for senior positions you don't have enough experience. It's a dilemma once you get into that situation.

Are you sure you can't still get into junior positions to gain experience with a PhD? I know that might be true for government jobs, because they have this salary tier that is dependent on both experience and education, but for industry jobs, I would imagine one could still get a junior position to gain experience with a PhD. It's like the over qualified myth for college grads. I think it's a myth, because I have been interviewed for really low paying jobs and even for a fast food position with a college degree on my resume. When I told an interviewer over the phone that I was in the processes of getting a master's degree, her response was 'Ooooo!' She wanted me to come in for an in person interview, and this was for a 12 dollar an hour job.

There are some jobs in which a PhD may make you more competitive, and there are a few in which it might even be essential, but for me, I decided just a few years ago that it was not worth the pain. However, some people actually enjoy the extended school thing and maybe it's not painful for them. For me, it would be more suffering, so there's no way I could do it. I'd rather be broke and out of school than broke and still in it.

 

[CrysPhys]

Here's an excerpt from a response I posted in another thread (https://www.physicsforums.com/threads/opportunities-for-an-aspiring-physicist.968038/). It's specifically geared towards a PhD in physics, but also applicable to other fields (primarily in science and engineering), in which you receive a full tuition waiver and a full financial stipend for your PhD program:

(b) Here’s the perspective I offer students that I’ve mentored. Unlike a degree such as a MBA, MD, or JD, a PhD in physics is not necessarily a means to an end: it can be an end in itself. Typically (in the US), you can earn your PhD debt-free: any university that really wants you will offer you a full tuition waiver plus financial support (e.g., fellowships, teaching assistantships, and research assistantships) sufficient to cover your living expenses (if you are single). You are free to pursue your research passion as a grad student (and as a postdoc if you wish). Once that phase of your life is over, you move on.

There have been multiple threads (including the one cited above) that discusses options for what you can move on to and backup plans for making the transition easier.