Your advice seems to be strongly biased by your experiences and several of your "facts" are simply wrong in many fields.
> Because for most people going into a doctoral programme is a bad idea. If you do not come from money or have a spouse who can support you it's a terrible bet in terms of chances of success.
In the hard sciences you get a pretty decent salary (scholarship). I know that in many countries in Europe they are significantly above average. That doesn't mean you should do a PhD for financial reasons (there are definitely better paying jobs out there), but I know of several people who support both kids and a spouse on a PhD salary, so you don't need to rely on someone else.
> Outside demand for PhDs is low in most fields.
That depends a lot on field. For CS this might be true (although quite a few jobs in ML expect PhDs for example). In chemistry and biology a PhD is almost expected if you want to work in industry. In my area (photonics/telecom) the vast majority of R&D roles are PhDs. Masters or bachelor degrees essentially get you a technician role.
>Most people who start a PhD don't graduate with one. They drop out.
I would like to see some stats on that. I know that in my experience (having seen more than 100 PhDs graduate) almost everyone graduated (I know of one who didn't, and he ended up with a very good job anyway)
> Substantial numbers of people who do get a PhD never even get a tenure track job. Getting a tenure track job often leads to looking for a new job after seven years, with the "rejected" sticker.
The first part is certainly true, becoming an academic is one of the less likely outcomes. However, regarding tenure track, the rejection really only happens at the most prestigious institutions in the US. In Europe this is not even a thing.
> Many fields are worse than that. They demand postdocs. Some people make the terrible mistake of thinking that adjuncting is a way of staying in the academy. It's not. If you are adjuncting full time you are extremely unlikely to ever get a real academic job, as a lecturer or professor. Get out.
I'm not sure I understand what you mean by adjuncting and what it's relation to postdocs is. Yes many fields require postdocs if you want a tenure position. Pay is highly dependent on country, e.g. in Australia you get paid on the same scale as other academics, in the US the pay is sometimes worse than a PhD.
> Getting a PhD is fine if you are in a field with great outside options, like Economics, CS, Math etc. but it's consumption, not an investment in future earnings. Outside top programmes the likelihood of becoming an academic drops of f rapidly. A PhD is training to become an academic. It can also be necessary to get a job in some fields with huge oversupply of PhDs, like the sciences, but those jobs are badly paid too.
It really depends what your life goals are. If money is your primary motivator, a PhD is not a great investment, do finance, programming or get into sales quickly (a good sales person can outearn all of the others). If you want to do R&D in many fields you require a PhD and pay can still be excellent. I know several people who were offered >150k jobs out of their PhDs.
> If you really, really want to be an acedemic and you can get into a good programme go for it. No one else should go unless they don't need the degree to earn a living or the degree comes with the prospect of radically improving their lives by making immigration possible.
This advice again highly depends on your motivation. If you want to just earn as much as possible as quickly as possible, yes don't get a PhD. If you are interested to work on hard problems relatively independently, a PhD could be right. One thing to realise is that it does require a high frustration tolerance, because there will be significant setbacks. That said everyone considering a PhD should do their homework on the group they want to join. Unfortunately there are bad supervisors, who consider PhDs like slave workers, or other groups who don't get you the publications that you should get, because the do poor research.
My SO started a PhD program two years ago. We had to move across the country for it. Since then their PI has moved to another university in another state, so we had to move again as well. The new university is in a city. Our rent is $1900. Their income is $29,000 / year. They work 40-60 hour weeks. And this is for a paid RA position. The work of these students is what the group lives off of. It's sustainable for the institution, but it's awful and needlessly burns out driven and capable contributors. That's a standard case in academia.
Also as an MS research DSP engineer: I work with PhDs who know far less and are far less productive than me. In some fields, for industry, it may really be just a piece of paper.
>> Most people who start a PhD don't graduate with one. They drop out.
> I would like to see some stats on that.
I was wrong. Over 55% of students get a doctorate within ten years of beginning. [Edited to correct error]
"the completion rate ten years after students begin their doctoral program remains low at 56.6% (Sowell, Zhang, Redd, & King, 2008). Additionally, the analysis indicates that completion rates continue to vary considerably by field of study: 49.3% in humanities, 54.7% in mathematics and physical sciences, 55.9% in social sciences, 62.9% in life sciences, and 63.6% in engineering."
> I was wrong. Over 45% of students get a doctorate within ten years of beginning.
I think the GP's point was that you're wrong on many fronts, simply because you've made broad, sweeping generalizations that don't necessarily apply to all fields, or even to any particular field across countries. Note even that this study you've cited is old and of very limited scope.
GP said in his experience "almost everyone graduated". If 56% graduate in 10 years that's far from everyone.
If you think I'm wrong on any particular I welcome correction. Feel free to cite some evidence that doing a PhD is a good decision for anyone who doesn't want to be an academic and doesn't view a PhD primarily as consumption.
I was merely indicating that the person to whom you initially responded (cycomanic) offered an excellent, and very detailed, point-by-point rebuttal of your initial broad comments about PhDs. I thought it provided a more nuanced explanation of how the overall training and career prospects vary by _location_ and _field of study_.
Thanks you said it better than I. I don't think my experience is necessarily representative either, but it clearly shows that in my area his statement is not true.
> I know that in my experience (having seen more than 100 PhDs graduate) almost everyone graduated (I know of one who didn't, and he ended up with a very good job anyway)
I know that's probably not what you meant, but saying 100% of PHD graduates graduated is the text book definition of survivorship bias. Obviously, if you look at a group made of only people who got a PHD, 100% will have a PHD.
I imagine you meant you knew 100 people who started a phd and not phd graduates?
I see how this was ambiguous. You're correct I meantof the more than 100 PhD students (i.e. students who started) I saw, essentially all but one graduated.
During my time in a computer science PhD program at an American university (2003-2010), my stipend maxed out at $23k a year. That was not competitive compared to having a normal job, nor would I have felt comfortable supporting anyone else.
Adjuncts professors teach at American universities on a course-by-course contract basis. These people usually have PhDs. Some would like tenure track positions. But it’s generally a career dead-end.
I suspect your experience is from Europe, and I don’t think it generalizes to the US.
But I never meant to say my experience generalizes. I simply meant my statement to provide a counterpoint, that the points brought by the OP are not generally true.
I agree that the US "in general" (again simplified) seems to be worse regarding PhD pay, but AFAIK there are significant variations across fields even in the US (i.e. engineering PhD students get paid significantly better than most others for example).
Yes, engineering PhD students get paid significantly better in the US. It's still very low compared to what people can get with a normal full-time job, and few people would be comfortable supporting anyone but themselves on it.
> Because for most people going into a doctoral programme is a bad idea. If you do not come from money or have a spouse who can support you it's a terrible bet in terms of chances of success.
In the hard sciences you get a pretty decent salary (scholarship). I know that in many countries in Europe they are significantly above average. That doesn't mean you should do a PhD for financial reasons (there are definitely better paying jobs out there), but I know of several people who support both kids and a spouse on a PhD salary, so you don't need to rely on someone else.
> Outside demand for PhDs is low in most fields.
That depends a lot on field. For CS this might be true (although quite a few jobs in ML expect PhDs for example). In chemistry and biology a PhD is almost expected if you want to work in industry. In my area (photonics/telecom) the vast majority of R&D roles are PhDs. Masters or bachelor degrees essentially get you a technician role.
>Most people who start a PhD don't graduate with one. They drop out.
I would like to see some stats on that. I know that in my experience (having seen more than 100 PhDs graduate) almost everyone graduated (I know of one who didn't, and he ended up with a very good job anyway)
> Substantial numbers of people who do get a PhD never even get a tenure track job. Getting a tenure track job often leads to looking for a new job after seven years, with the "rejected" sticker.
The first part is certainly true, becoming an academic is one of the less likely outcomes. However, regarding tenure track, the rejection really only happens at the most prestigious institutions in the US. In Europe this is not even a thing.
> Many fields are worse than that. They demand postdocs. Some people make the terrible mistake of thinking that adjuncting is a way of staying in the academy. It's not. If you are adjuncting full time you are extremely unlikely to ever get a real academic job, as a lecturer or professor. Get out.
I'm not sure I understand what you mean by adjuncting and what it's relation to postdocs is. Yes many fields require postdocs if you want a tenure position. Pay is highly dependent on country, e.g. in Australia you get paid on the same scale as other academics, in the US the pay is sometimes worse than a PhD.
> Getting a PhD is fine if you are in a field with great outside options, like Economics, CS, Math etc. but it's consumption, not an investment in future earnings. Outside top programmes the likelihood of becoming an academic drops of f rapidly. A PhD is training to become an academic. It can also be necessary to get a job in some fields with huge oversupply of PhDs, like the sciences, but those jobs are badly paid too.
It really depends what your life goals are. If money is your primary motivator, a PhD is not a great investment, do finance, programming or get into sales quickly (a good sales person can outearn all of the others). If you want to do R&D in many fields you require a PhD and pay can still be excellent. I know several people who were offered >150k jobs out of their PhDs.
> If you really, really want to be an acedemic and you can get into a good programme go for it. No one else should go unless they don't need the degree to earn a living or the degree comes with the prospect of radically improving their lives by making immigration possible.
This advice again highly depends on your motivation. If you want to just earn as much as possible as quickly as possible, yes don't get a PhD. If you are interested to work on hard problems relatively independently, a PhD could be right. One thing to realise is that it does require a high frustration tolerance, because there will be significant setbacks. That said everyone considering a PhD should do their homework on the group they want to join. Unfortunately there are bad supervisors, who consider PhDs like slave workers, or other groups who don't get you the publications that you should get, because the do poor research.