If you were a professor, how would you choose your PhD students?

<p>I think about this, and it worries me. I’d like to be a professor. I’ve seen so many situations were profs make bad choices as to who to take as students. This is in engineering, so students can drop out with a masters degree. Professors fund students who leave. What qualities do people think make the best PhD students? How do you find out if a student has these qualities? </p>

<p>As a prof, idealistically, you’d want to chose the most accomplished and most deserving students. At first I’d likely look at this differently though…I would choose students so as not to make mistakes. I want them to fit into my lab and make it thru the program.</p>

<p>My major concerns are:</p>

<ol>
<li>I think undergrad grades are less than a mediocre predictor of PhD success. I know I’m not alone in saying this. Almost every prof who’s talked about how to hire PhD students has said the same thing. There are multiple problems with undergrad grades.
1st: freshman and sophomore grades are largely dependent on a student’s high school education or whether the student has their act together as a freshman.
2nd: junior or senior classes are often either graded high, or at some smaller schools grades are given out “politically” to position certain students for certain grad schools.
3rd: Grad course are a whole different ball game. They are very heavy on project and group work compared to undergrad classes
4th: Classes only make up 1/3-/4 of your PhD career. The rest is research.</li>
<li>The GRE is useless. It tests high school math for graduate level admission. There are no subject tests for engineering.</li>
<li>An engineering masters degree is a marketable and respectable degree. Students can take it and run. Sometimes the best qualified students take it and run. Sometimes students aren’t up front about going into grad school only wanting a masters degree.</li>
</ol>

<p>So how else would you choose PhD grad students? Comments and thoughts would be appreciated</p>

<p>Well, I am not in an engineering department but I am an experimental materials physicist and I have had a number of Ph.D. students over my years at Illinois Tech. I have had some students who have left after a Masters and that is simply an occupational hazard. In my department we usually take on students who have passed their qualifying examination and have done well in their first year(s) of coursework. Occasionally, I take on a Masters student but usually do not provide support.</p>

<p>I generally look at the students record and speak extensively with them to figure out whether they would fit in the group well and how serious they are about a Ph.D. I try to make a go/no go decision within a semester and if they can volunteer while they are still Teaching Assistants, that helps too.</p>

<p>You never know if a student will become disillusioned halfway through but the best way to make sure this is lower probability is to try to keep the number of years in the program to 5 or less from the B.S. degree. That means managing their project a bit more if the student is floundering and pushing them to keep a schedule on required examinations.</p>

<p>xraymancs, what you described is what I’d like to do; however, programs, like umich where I graduated from, fund nearly %100 of the students in their PhD program. Profs often pull students the masters program at umich, but many of the best students will not accept Masters admissions or unfunded PhD admission. A very very common discussion on College Confidential is “How do I get funding?”.</p>

<p>This is where the trouble lies…students accepted into the PhD program who do not make it thru and leave with a masters degree.</p>

<p>It’s hard to tell whether there are people who lie on their personal statement prior to admission, since the personal statement can be faked to some extent. There are students who say in their personal statement that they want to do, say, research in an industrial setting with their PhD, but who really just want a masters from that department because they realize they really want to work an industrial job that only requires a masters, for example. </p>

<p>Some internationals leave with a masters so that they can go to another school to do a PhD for real because, presumably, PhD-level jobs in their home countries prefer, or even require, both a MSc and a PhD; India is notorious to require both a masters and a PhD for university faculty jobs, or to otherwise assume that a PhD holder also holds a MSc.</p>

<p>Their past research track record definitely holds weight in my decision, were I a prof that had decisions of that sort to take. And I would take the PGRE with multiple grains of salt.</p>

<p>If the department does not give TA support to first year students but relies on RA positions for all of them, then there is a risk and it is often why such departments expect a student to already have a masters before applying. In physics this is usually not the case and students in a Ph.D. program will generally have one to two years of TA position guaranteed by the department. This give faculty some time to assess the students and those who want to leave with a Masters will often just take the coursework-only degree after 2 years.</p>

<p>Bottom line, students change their goals and it is always possible that they might leave with a Masters only (I just had one student do this in May). It’s just part of the job…</p>

<p>^I am with @xraymancs. I have a PhD but am not yet a PI (I am a postdoctoral researcher). As a PI, you always run the risk that a student is going to leave the program before completing the PhD. Some students may plan this ahead of time and “pretend” they want a PhD when they really only want an MS. But I’m willing to bet that the majority of students who leave simply change their priorities or get burned out or realize they don’t need a PhD to do what they want to do. When you fund a student you know there’s always the possibility that they will leave before finishing - just like when you hire an employee you always run the risk that they will not be a good fit and leave before you can recoup the costs of hiring them (or worse, that the employee will cause trouble).</p>

<p>With that said, my main concerns in selecting a PhD student would be 1) research fit, 2) research experience, 3) academic foundation, 4) intellectual curiosity and 5) personality fit. I would want a student to </p>

<p>1) be excited about the research being done in my lab, and see connections between her own interests and the work that I do. This increases the likelihood that she will stay in grad school and stay working with me. This comes from the personal statement and interview.</p>

<p>2) be experienced. First, at the graduate level you expect not to train doctoral students to do basic things. But secondly, research experience gives me more of an assurance that she knows what she’s getting into and won’t bail on me because she discovers she hates research. I would not take on a doctoral student who didn’t have any research experience, and I would give preference to students who would have at least 2 academic years by the time they joined me (so for someone straight out of undergrad - you started junior year, and maybe had a summer REU).</p>

<p>3) have a good academic foundation, so I don’t have to waste time teaching her things she should’ve learned in undergrad. Graduate school research and coursework builds upon a theoretical foundation that you acquire in classwork, so while grad school is not about coursework and is more about the research, the coursework is still a really important foundational part of the program. This is where college GPA comes in. I disagree with your assessment of GPA, OP; I don’t think “political” awarding of grades is widespread enough to be a concern. And yes, HS education may influence freshman grades - but while I am sympathetic to that, I want someone who can help me do the work that needs to get done and that will succeed. If you have to take remedial math freshman year because your HS education is terrible - and do poorly in it - that’s sad, but what is my assurance that you can correctly do advanced statistical analyses when I need you to in the lab, unless you show significant improvement later on? By sophomore year you’ve had some time to adjust, so the adjustment excuse doesn’t hold as much water.</p>

<p>4) be intellectually curious. I don’t think that grades are the end-all, be-all, nor do I believe in hard cumulative thresholds. I would rather have a 3.2 who is intellectually curious, independent-minded, and generates new ideas and sees interesting connections than a 3.8 who only knows how to regurgitate information to me. Heck, I’d rather have an independent-minded, thoughtful, brilliant student who had a 2.6 in undergrad but has otherwise shown that they can complete graduate-level work than have a 3.8 who is going to annoy me with elementary stuff because she didn’t learn how to go beyond retaining facts in undergrad. I would look for upward trends and major GPAs, and take into account extenuating circumstances, especially time since the BA. I don’t care at all about GRE scores, as I think they are rather meaningless - and I say that as someone who scored very highly on the GRE. Grad school is all about pushing knowledge forward, but simply spitting out what you read in a book.</p>

<p>5) fit with me personally. Many times when students contemplate leaving, it’s because of a personality mismatch with their advisor. They want mentorship in a style that their advisor is unable or unwilling to give. Like me, I am not a micromanager. I need grad students who don’t wait on me for every little instruction. If I had a grad student who needed my go-ahead before they did anything, she would likely be unhappy because that’s not the way I work.</p>

<p>And how often would people “fake” their desire to earn a PhD and really just want a masters anyway?</p>

<p>

</p>

<p>More often than you think. Engineering is very lucrative. </p>

<p>I don’t think it happens terribly often, but I also don’t think it’s rare. Master’s degrees are less often funded than PhDs, and it’s possible to get into a PhD program and get a free MS out of the deal, and then leave. I have had students in fields other than engineering ask me whether or not they should do this. I always tell them no - I mention the ethical concerns but also focus on the practical concerns, which are that PhD programs are geared towards preparing researchers and that you won’t get the kind of professional development and advising that you would get in a professionally-oriented MS program (not to mention that the MS can take longer than the standard 2 years if your advisor is not really concerned about you earning the degree. It took me three years to earn my MA on the way to my PhD, because I was trying to publish my master’s essay and so took longer to write it, but I didn’t care because the end goal was a PhD). Plus, to me it would seem pretty miserable and torturous to be in a PhD program for two years if your end goal wasn’t the PhD.</p>

<p>But the prospect of saving tens or hundreds of thousands of dollars can cause people to overlook that, I suppose, or think it’s worth it.</p>

<p>I don’t think that it’s ethical to enter a PhD program and take your professors research grant money with the intent on leaving after a Masters, but once you’re in a PhD program, it can be a long and dreary road. If you find that you have lost the fire in the belly after the Masters, I think it’s perfectly reasonable to quit and get a job. Also, not everybody passes their qualifiers. </p>

<p>No doubt getting a PhD can be a long, frustrating, and dreary road.</p>

<p>I don’t think that many students are downright unethical. I think few students say to themselves “My goal is to get funding meant for PhD students, and run away with a masters in engineering.” Instead they are still deciding on their life goals and not writing that in their applications. Then they are taking funding. They think they can walk away with a Masters degree if they lost their fire as ClassicRocker just said. </p>

<p>I wouldn’t want to fund or work with this type of student. I would want to root them out. They are especially damaging to the careers of assistant professors. They should enroll in a Masters program and pay for it. Part of my question is how do you read between the lines in their application to find this out? How do you interview them to find this out?</p>

<p>In engineering this happens with about 20-30% of American students, and perhaps 10% of international students. Much of the fellowship money available for 1st and 2nd year students in engineering is only available to Americans.</p>

<p>What about other STEM disciplines?</p>

<p>My son was honest in applying only for his master’s. We have both heard and read over and over again that grad school funding is generally for PhD candidates and does not generally go to master’s candidates. But, of course, we’re both hoping that he will be an exception to the rule.</p>

<p>There is a professor who has shown a lot of interest in him for some time now and has urged him to apply to grad school.</p>

<p>Please forgive what I assume is a very naive question. … If a student is a desirable candidate, and the student shows promise as a researcher and/or as a grad student, then why would a program or professor not want to fund that student for a master’s? Why is it more desirable to fund a student for a PhD?</p>

<p>What is the major purpose of funding a grad student anyway? I’m pretty sure I understand why undergrads are “funded” (given massive scholarships), but why doesn’t the same reasoning apply to all grad students, regardless of whether they are getting their master’s or their PhD?</p>

<p>A prof has limited money to fund students, and they want a return on their investment. Also, often a metric for a prof receiving tenure is how many PhD students have graduated under that prof. </p>

<p>To fund 1 grad student for 1 year from an external grant the professor applies for (i.e. NSF, DARPA, NIST, etc) costs ~$75,000. To begin with the university takes a certain percentage of the grant (e.g. 40%). The rest of the money is used to pay the students tuition and stipend.</p>

<p>I took me 2 years before I had enough experience to really start doing independent research as a graduate student. My first year was a fellowship, and then my advisor covered the rest of my PhD. He had to invest in me before I learned enough to start to be able to do meaningful research.</p>

<p>To get another grant, profs need to produce results on the first grant they applied for too. They need competent grad students. Also, it may take an experienced professor ~1 month to write a grant where he might get ~$400,000. This may cover research expenses and the funding of 1 PhD student’s time at a university along with other research expenses. Only ~20% of grants get accepted from a top ~20-30 engineering university.</p>

<p>FYI: Masters students can get funding in some ways other than profs paying thru their grants.</p>

<ol>
<li>National labs or companies have programs to pay for a masters education full time as a recruiting tool…or for their employees</li>
<li>Occasionally, to recruit great masters level students, the school will offer the chance to be a TA to get full funding.</li>
<li>Sometimes a TA position is available your 2nd semester in grad school. I was able to be a TA with full funding after my first year in my terminal masters program. This would be unheard of where I got my PhD at Michigan though.</li>
<li>Sometimes profs will pay students hourly, and they will receive the equivalent of a stipend. They will still need to pay tuition.</li>
</ol>

<p>^^ I didn’t know there was a tenure metric based on how many PhD students have graduated under a prof. That’s one good reason why the PhD students might be more “valuable” to a prof or department, and hence why a PhD student might be more likely to get funding over a MS student.</p>

<p>But other than that reason, are there other reasons why a top-notch, competent applicant who only wants to earn a master’s wouldn’t be on equal footing with a top-notch, competent applicant who wants to earn a PhD?</p>

<p>I guess what I’m asking is why wouldn’t a fresh-from-undergrad master’s applicant and a fresh-from-undergrad PhD applicant be on equal footing for funding potential? Aren’t both students going to be doing research and starting from the same place anyway?</p>

<p>Is there something else the department gains by luring a PhD applicant over a Master’s applicant? Is it as basic as the amount of time that the student will be there – like preferring to hire an employee who agrees to stay for 4 years over one who agrees to stay for 2 years? So they can save themselves the trouble of “hiring” and “training” a new person 2 years later? Is that why PhD’s are more fundable?</p>

<p>I’m trying to learn more about it. I got my own master’s degree years ago. I was not funded. Strangely, it didn’t even occur to me to ask for funding back then, and I didn’t feel disappointed that I didn’t get it. Ignorance is bliss! :slight_smile: </p>

<p>At an engineering research university (top 30), my understanding is that the criteria to earn tenure it breaks down as following…</p>

<p>1/4-1/3 on teaching evaluations
The rest is based on research. They key metrics for research are…
-The most important thing is lots of published peer-reviewed journal papers…10-15 in well respected journals. The process of doing the research, writing, submitting to the journal, rewriting the requested revisions, etc. for a single journal paper usually takes 3 years. A masters degree usually is 2 years.
-1-5 students earn PhDs under your guidance
-You’ve brought in lots of grant money >$500,000 - $1,000,000
-You are respected in your field. (e.g. people know about your research from conferences you or you students attend and your publications). External letters of recommendation are critical for profs getting tenure.</p>

<p>That will do it. You need highly trained grad students to write papers. One professor can’t write 10-15 quality journal papers in 6 years. You have to work with well trained grad students that you supervise. To get and keep grants you need results. Again you need well trained grad students that stay for 4 or 5 years to get results…so yes it has to do with “training” grad students, and how much time you get from them once trained.</p>

<p>FYI: I prefer to think of grad students as “apprentices” or “faculty in training” rather than “employees”. I company doesn’t necessarily care what an employee learns, but a university absolutely does care what a PhD student learns. A PhD student doesn’t work 9-5 and should not be expected to be constantly supervised…an employee is basically constantly supervised during a set time every day.</p>

<p>^ I wasn’t saying grad students were employees. I was comparing hiring of employees to choosing amongst applicants for funding. Based on what you’re saying, it sounds like that’s probably the main reason behind funding PhD’s over master’s candidates. Most master’s students can’t turn out a product for the professor in just 2 years, so the prof doesn’t get the payback for doing all that “training” if the student finishes and leaves in 2 years. </p>

<p>(FWIW, though, most companies do care what an employee learns and how fast he or she learns it, and employers tend to hope that employees won’t need much supervision, as well.)</p>

<p>@SimpleLife - A Ph.D. student will stay on a project for about 3 years and a Masters student will only have about 1 year (the coursework is the major part of 2 years). If I have a grant that has to produce results and every year I am retraining a new Masters student who never gets to the point of publishing (or just one publication), I will prefer a Ph.D. student who become more productive in the second and third years and then can help train new students. Of course a good Masters student is welcome but an equally good Ph.D. student is better.</p>

<p>Graduating Ph.D. students is important for getting tenure but more important than that is publications and Ph.D. student (and postdocs) are more effective at that.</p>

<p>I think trying to root out PhD students who are still deciding on their life goals is a fruitless exercise, since the vast majority of PhD students are doing that and even the most serious and dedicated PhD student is probably not 100% sure what they want to do with the PhD yet. I was a pretty successful doctoral student - externally funded throughout, including an NSF; several publications; got a postdoc a year before I graduated; and my advisor got tenure in part because of my role in his lab - but I went into the program still deciding on my life goals and realizing that I could walk away with an MA if I didn’t want to finish the program. That’s just reality. You always need to have an exit plan.</p>

<p>Like I said, going to a PhD program with the plan in mind to definitely drop out after you get the MA is completely different than going in with every intention of finishing, but with the knowledge that you can drop out if you decide that the program is not for you. The latter is simply knowledge; it’s like knowing the sky is blue or that bears eliminate waste in the woods. It’s simply a fact.</p>

<p>

</p>

<p>As for why PhDs and not MAs, it’s mostly time. New graduate students are much less useful than advanced grad students, unless they have extensive research experience/history. Even then - you have to learn the way the lab works, you have to learn the new data set or experiments, new equipment, perhaps some new analysis methods, new protocols. You have to get on IRB or IACUC protocols. You also need to acquire soft skills, like how to write scientific journal articles (as most have never written one before PhD program), how to assist in grant-writing, and how to present at conferences. There’s a lot of ramp-up time. Also, during most of this time you are taking classes, which diminishes the amount of time you can spend in the lab.</p>

<p>I would say that it probably takes a new grad student at least 1.5-2 years to really get up to speed and to the point where they can work efficiently and semi-independently, without time-consuming guidance from a professor. Dissertation phase PhD students in years 4 and beyond are even better. In fact, I would say that a senior undergrad can probably do most of the same tasks as a first-year graduate student, and they are much cheaper (most of the times free!) If you’re an MA student, by the time you’re more useful and have more available time than a free undergrad, you’re finished and out the door. But as a PhD student, you still have another 3-4 years for the professor to get your assistance. Also, by year three you are largely done with classes, which means you can spend almost all day in the lab if you want (and the professor no longer needs to pay your tuition, which they would if they had to hire another MA student to replace the MA student they just lost).</p>

<p>Not only that, but academia is a slow-moving beast. A grant can take 2-3 years to get funded from conception to grant-writing to acceptance to funding. Projects themselves can be 3-5 years long - it might even be 1-2 years before you have enough usable data to do useful, interesting analyses and get publications. Professors want students who are going to be there long enough to understand the project, get really familiar with the work, and be able to churn out those papers when it happens. If your advisor is just collecting data in your first year of an MA, you might be gone before he starts to write papers. Then he needs another student to help him.</p>

<p>Also, PhDs contribute to the prestige of universities and departments. Universities are more prestigious, and ranked higher, with PhD programs. This happens both directly (the actual existence of the program = higher ranking) and indirectly (PhD students help write grants that bring in more money; they help write papers that increase the visibility of the institution; they hobnob at conferences and give presentations and teach classes and all that). PhD students also graduate and take prestigious tenure-track jobs, which increases the visibility of the program. MA students do that stuff far less frequently - or, in the latter case, not at all.</p>

<p>Postdocs can come on for 2-3 years only because they have the PhD, and so they’ve already got the training. Ideally they’ve already helped co-author some publications, seen the way a grant works, and done some presentations. They can kind of hit the ground running with the postdoc begins. They only need to learn about the project, not learn about how to be an investigator period. Plus postdocs don’t have any distracting obligations to a graduate program. I’m a postdoc now and it’s like heaven to do research all day without worrying about my dissertation or studying for exams or whatnot.</p>