<p>I do agree, though, collegealum – if a professor is willing to go to bat for you, that’s gold. The programs I know don’t have you apply directly to a lab, but if a professor is interested in recruiting you as a student and is either on the admissions committee or communicates this desire to the admissions committee, then you will almost certainly be admitted. </p>
<p>
Should a faculty member value the opinion of a colleague (even one outside his/her specific subfield) on the technical aptitude and preparedness for graduate study of a student the colleague knows well? Absolutely. I don’t think an evaluation of an undergrad by a professor requires intimate technical knowledge about the work the undergrad intends to do in graduate school.</p>
<p>Well, in so much as they can comment about the student’s potential for success clearly and meaningfully yes. Maybe I was unclear - why does knowing a colleague well as opposed to knowing a professor, colleague or not, is qualified to comment intelligently about a student, make a difference? And should it?</p>
<p>I think knowing a colleague personally makes a difference in how the letter can be interpreted – it makes the letter more informative.</p>
<p>I heard from a few professors on the interview circuit that my rec letter from my undergrad PI was a great letter, but it was remarkable because they knew that, for him, it was a particularly enthusiastic letter about a trainee.</p>
<p>OK, you’re right, I would agree it’s unreasonable to deny there’s no context to be gained from knowing recommenders personally. Still, my impression is strongly that professors who do admissions tend to have some experience doing it, know a lot of the top scholars, and probably have seen letters from them - I gather this impression from speaking to a few of them (but of course, my experience narrows on a field…and here it really may make a difference, since the skills commented on for my field are described very much in the abstract). Usually it’s very clear what the recommendation means. Perhaps this is why in theoretical field, the ‘home school advantage’ is quite limited. </p>
<p>Admittedly the interest of many reading this may not be theoretical; still, I felt the reason some of the things being thrown around are true is better said than not.</p>
<p>I disagree; as far as I’ve seen, that seems more common at MIT. What ‘other places’ are you talking about—other institutes of technology?</p>
<p>
</p>
<p>I can see the master’s—Stanford has a coterminal program where students can get their master’s and bachelor’s in five years, which you could argue is an “institutionalized” preference for undergrads in its master’s programs. However, my understanding is that MIT does show some preference for PhD programs; sorry if that wasn’t clear in my first post. </p>
<p>molliebatmit,</p>
<p>
</p>
<p>That would make sense, but two professors at MIT told me explicitly that they look more favorably on MIT undergrads, which suggests that it’s more than just a shorthand.</p>
<p>
</p>
<p>Oops, yes, I misspoke—I wasn’t trying to imply that MIT cares more about background/representation than research/recommendations. Rather, I’m saying that MIT doesn’t care about background unless that background includes a bachelor’s from MIT. Of course that’s probably not 100% true, but the point stands.</p>
<p>
</p>
<p>Right, I agree that students at other schools who have access to these opportunities would somewhat have an unfair advantage over those who don’t. However, I wasn’t trying to indict the admissions game in general, but rather that, of the students who attended universities where these opportunities are equally plentiful, those at MIT will still have an advantage. Assuming that MIT does in fact show an explicit preference for its undergrads, one of the following is true: MIT realizes that those at other schools are usually prepared with the same opportunities but gives preference to MIT students anyway, or MIT truly believes that its students have superior opportunities/preparations (this is where the ‘arrogance’ part would come in). </p>
<p>Of course these would be generalizations— “MIT” isn’t the one to blame, but the individual programs, which may vary wildly in their preference to MIT undergrads. And I could be completely off-base in saying that MIT is arrogant by assuming that its undergrads will usually be better prepared—perhaps they are. But I have difficulty believing, no matter how many times people on this site will tell me, that MIT is superior to Stanford, Harvard, etc. in providing its students opportunities like research and internships.</p>
<p>
</p>
<p>I agree that it’s not likely that two applicants are equivalent, but that MIT would show any preference to its own undergrads for CS over students from CMU is just… absurd. To me, at least.</p>
<p>By the way, my diatribe, as college314 would put it, is not out of bitterness, as I was recently admitted to the PhD program in CS at MIT, but I can’t help but think that had my recommenders not been more renowned than the professors I’m interested in at MIT, I wouldn’t have gotten in, even though I’m easily as qualified as MIT undergrads in CS. The way the MIT professors in CS had told me, I thought I wouldn’t get in because of the preference to MIT students, even though Stanford’s a top school for CS. That was rather frustrating in the process, especially since it turned out to be just all scare (what’s more annoying is that one of the two professors I was interested in left MIT for an Ivy a few weeks ago, so that kind of throws some of my plans out the window; really wish both professors would’ve been more up-front about it, since this wasn’t unforeseen).</p>
I would think the first possibility would be more arrogant than the second – presumably, if a department believes its undergraduates are best-prepared for graduate study in the department, this comes from data collected over the years: graduation rate, time-to-degree, jobs landed after graduation, etc.</p>
<p>Using general trends to justify a decision for a specific candidate without providing explanation for the general trends is also dangerous, or at least it would seem to me. That is, even if it seemed statistically that MIT undergrads fared better in MIT grad school, it would be important to ask why. </p>
<p>The sense in which the first option is less arrogant is that it acknowledges there isn’t a good reason. Making a claim that may be disputed, without offering a good reason, can also be a big form of arrogance.</p>
<p>And given their day jobs as world-renowned faculty members in technical fields, I would also assume that this data, if collected, has been analyzed ten ways to Sunday. </p>
<p>Particularly in some of the engineering departments, MIT undergrad education is different from (and more intense than) is typical. In aerospace engineering, for example, there’s a giant concatenation of a course called Unified Engineering taught to sophomores, which contains class material that most aero/astro programs distribute over two full years. If the department has gone to such lengths to design such a course, it doesn’t really surprise me that they feel their own undergraduates are best-prepared for further study in the department, and it doesn’t strike me as arrogant, either. If (e.g.) Stanford decided that the approach exemplified by Unified Engineering didn’t produce students who were successful in its graduate program, I would not find that arrogant, either.</p>
<p>I have no doubt some programs are more intense than others on average, but still, candidates are specific people. So ultimately one has to read the application and decide if comparable preparation was achieved or not. I would have no objection if a more intense program (on average) has more successful applications. However, that is very different from explicitly preferring MIT students – more precisely, I would hope it’s preferring the level of preparation (which, I would think, can be achieved at a top school elsewhere … but maybe not).</p>
<p>I’m sure that MIT has analyzed all that data to death, but I’m just going off the assumption that even such data (graduation rate, time to degree, etc.) reveals no difference between undergrads from MIT and those from, say, Stanford or Harvard. If it doesn’t, then that would be justification for saying MIT’s practices are unfair. My assumption might not be the reality, but if it is, an additional possibility is that MIT’s preferences vary not only by program but also by undergrad institution or by the prestige of the program of interest at that undergrad institution. Thus, it could be the case that MIT CS prefers its own undergrads and also gives a ‘boost’ to those coming from Stanford CS, Berkeley CS, etc. Of course I think it’s generally acknowledged that undergrad institution does play a role, to varying significance (though generally not very important), in graduate admissions at (almost?) every school, simply because they tend to know about the schools that have a higher-ranked program or are ranked higher on the whole, and because, as you say, they have lots of data on the different types of students they enroll and graduate. At this point there really aren’t any conclusions we can make, because we don’t have that data.</p>
<p>I agree, also, that the programs might have their own unique take on the discipline and that influences their undergraduate program (say, if a program required its undergraduates to undertake significant interdisciplinary education/research, which was also present in the graduate program); that would be justification to prefer their own undergraduates. But my intuition is that the material is the material, and Stanford CS, for example, isn’t going to have any knowledge that, say, UC Santa Cruz CS doesn’t, so what ends up distinguishing them is what the students did in their time as undergrads, namely research, and that’s where the top programs tend to have an advantage. I haven’t been in the field long enough or had enough experience with different programs to say whether there’s a qualitative difference in how the field could be approached at another program, to the extent that it’d warrant special consideration for those who were trained under said program as undergrads.</p>
<p>It’s also very possible that the MIT professors I talked to were exaggerating, so my view could be unrealistically skewed (though had I been rejected, you can bet I wouldn’t be saying this).</p>
<p>It is true different modes of preparation can be useful for different programs. In particular, different programs can emphasize different parts of the application to varying degrees. It would still strike me as somewhat strange if a school accustomed its undergrads to a specific program style and coincidentally that style were considered ‘best’ preparation for its PhD program, based on a phantom set of data and justifications nobody knows about. What would help is if they were transparent and said in their site our program primarily selects based on the following criteria, and then interested students can tune their skills appropriately. I am myself also against the opposite scenario - where the school selects against its own students. There is too much variation in career paths for the school to make the decision for the individual automatically, without viewing his/her specific case, and providing an opportunity to close any gaps that may be caused by difference in program philosophy.</p>
<p>After all, it is highly conceivable that a student outside of school X may prefer the graduate philosophy of school X to that of Y, if the student only knew what this philosophy is.</p>
<p>I’m pretty sure plenty of Berkeley undergrads in CS go back to grad school there. Not sure if they explicitly prefer Berkeley undergrads to, say, Stanford/MIT/CMU undergrads, but they seem to be very happy to keep them part of their family at least. And I quite think significantly more so than in some other fields.</p>
<p>Interesting–I remember hearing the opposite about Berkeley (that it discriminates against its undergrads in grad admissions). Actually I’ve heard that all UCs tend to prefer students from other schools in grad admissions. Probably depends on the field/school, even the professors on the admissions committee.</p>
I certainly hope no one is even considering the argument that MIT departments aren’t even reading the applications of applicants from schools other than MIT.</p>
<p>I would argue that we see, on average, a large number of MIT students admitted to particular MIT graduate programs because the faculty members in those departments feel that those particular undergraduates have the best applications and the greatest likelihood of success in the program. The fact that they attended MIT is causative in terms of the particulars of their applications (the courses they took, the professors they worked with), but not in and of itself (“this person went to MIT? ADMIT”).</p>
<p>No professor I know would deliberately handicap his or her research program by admitting less-qualified students to his or her graduate program over students with superior qualifications. </p>
<p>
You are also assuming that the admission rate for undergraduates from MIT is actually significantly higher than the admission rate for undergraduates from other top engineering programs. I would not make this argument.</p>
<p>Very much what I think is true. My info about Berkeley CS is admittedly given from someone who is a current grad student there. It also seems to be what some EECS students who are undergraduates there believe.</p>
<p>I’ve heard of years when Berkeley math rejected 100% of the undergrad PhD applicants. </p>
<p>
</p>
<p>Rest assured, I’m not :)</p>
<p>
</p>
<p>I guess I am fine with all this as long as professors from MIT seem to favor the preparation of MIT undergraduates just the same as they favor the preparation of someone who achieved highly from a school of comparable repute for the field/subfield of concern. </p>
<p>What my previous post was about (and this may have been clear already) is that if MIT undergrad’s style explicitly tailors you for MIT grad school, for example in CS (in a way that, say, CMU, antoher top CS school, doesn’t), and professors strongly feel this, then I would think it best for them to make it clear what they seek when they read applications. I have heard in a field like EECS, admissions can be done on the basis of some professor literally picking up a student’s application, realizing that the student would like to work with him and saying OK. I have also heard of professors publishing what they would like from their students before they begin work with them. </p>
<p>Stated concisely, if there is to be no confusion that professors prefer students on the basis of brand name rather than on the basis of the preparation that brand name provides, then it should be made clear what is expected by that department.</p>
<p>I may be doing a terrible job explaining this, but hopefully it makes sense to someone.</p>
<p>
</p>
<p>Sure, I wouldn’t think so either. </p>
<p>Remember, I don’t even know that MIT offers any preference at all to its own undergrads. I am merely commenting on what people are saying.</p>
<p>As I said, I’d assume that they’re comparably low, but I’m not sure how different they are, and I’m sure MIT won’t tell us. ;)</p>
<p>mathboy98,</p>
<p>
</p>
<p>This, I think, is partly where my frustration was coming from–those expectations seemed unclear, as evidenced by the many variables we’ve discussed in this thread. </p>
I’d actually assume they’re comparably high – even for top graduate programs, it’s not that difficult to be admitted if you have a strong research background, good recommendations, and a good GPA coming from a school that’s strong in your field of interest.</p>
<p>I have trouble getting worked up over a lack of clearly-stated benchmarks for admission to a graduate program, to be honest. Maybe that’s my fifth-year-PhD-student jadedness kicking in.</p>
<p>I agree with Mollie. And I think that CS is the worst example to make generalizations about graduate school admissions. Of all the technical fields, I think CS is the hardest to reduce to metrics such as classroom performance. And without such metrics, it becomes more subjective by necessity.</p>
<p>@Mollie, what some of us meant was that if indeed there are aspects to MIT’s undergrad prep that are considered superior prep for its PhD programs as compared to what is offered at other top research schools, applicants would probably like to be alerted what MIT’s own program uniquely expects. If there are no such expectations, then there is no question.</p>
<p>Also, depending on the field in question, those 3 things may be uncommon even among students at top programs. So the question becomes how to maximize chances when not nearly a genius.</p>