Tangibles vs. Intangibles in graduate admissions

[ringo-ring]

From what I learnt here at this forum is that such application components as:

• letters of recommendation
• statement of purpose

- are the most important in the whole application, while more objective measures of performance such as GRE scores, GPA and publications are the latest admission committee wish to look on!

In my opinion, this is completely wrong, since LORs and SOP have much more to do with social skills than research ability.

To write a good SOP, you have to know how to convince the reader, how to form the opinion you need through writing.

To get a good LOR, you need to build a good relationship with recommender and somehow make him think good about yourself. It depends as much on the ability to present yourself in a shiny colors (which doesn't have to do anything with actual state of things) as on the ability to "befriend" the recommender, which as Dale Carnegie and many others has shown, boils down to a few simple things: be nice, hold the same opinion, and do not forget to mention he's a great scientist when talking to other lab members... and, avoid showing up yourself as too smart or independent, i.e. arrogant.

Of course, social skills are also important since you'll have to work with other people in a lab, present your findings, mentor undergrads etc. but should they be the deciding factor? All in all, science primarily is not about being nice, but about the passion for research.

[Strangefox]

To write a good SOP, you have to know how to convince the reader, how to form the opinion you need through writing.

I cannot agree with you here. I don't think that SOP is about convincing somebody, it's about showing yourself as a person, something beyond tangible scores, and telling about yourself, your potential research, your fit. That is why most people recommend writing a SOP using the easiest language possible. It's not your hooks and brilliant style that will make you stand out, but your fit as a person and as a researcher. If there is fit, they will want you, if there is no fit, they won't, no matter how exellent your rhetoric skills are.

And of course they need to see that you can express your thoughts clearly, otherwise how will you be able to write articles and to communicate your ideas to other scholars?

It seems that when you write about convincing, you refer to a negative connotation of this word, that is, making somebody believe something while this something is not exactly true. For example, convincing adcomms that you are a good scholar while you actually aren't. I, personally, see SOP not as convincing but as revealing, revealing your fit, your experience, etc.

Scores, on the other hand, do not reveal neither your research, nor your personal fit with a department, that is why they are less important.

[Kitkat]

I cannot agree with you here. I don't think that SOP is about convincing somebody, it's about showing yourself as a person, something beyond tangible scores, and telling about yourself, your potential research, your fit. That is why most people recommend writing a SOP using the easiest language possible. It's not your hooks and brilliant style that will make you stand out, but your fit as a person and as a researcher. If there is fit, they will want you, if there is no fit, they won't, no matter how exellent your rhetoric skills are.

And of course they need to see that you can express your thoughts clearly, otherwise how will you be able to write articles and to communicate your ideas to other scholars?

It seems that when you write about convincing, you refer to a negative connotation of this word, that is, making somebody believe something while this something is not exactly true. For example, convincing adcomms that you are a good scholar while you actually aren't. I, personally, see SOP not as convincing but as revealing, revealing your fit, your experience, etc.

Scores, on the other hand, do not reveal neither your research, nor your personal fit with a department, that is why they are less important.

I agree with that. I would also say to the OP that I don't think people here over all say that GPA, GRE and all of that don't matter, they do. It's just that for so many programs, most of the people applying to any given program, already have good GPAs, GRE scores, and so on that, that the deciding factors come out being the SOP and the letters of Rec, if just to see if that person 1) will actually be able to do grad level work as judged by other PhD's and 2) if you would actually fit well in the lab. Just because you have a 4.0 GPA and a 1600 on your GRE does not mean your as good a fit for that lab as some one with a 3.7 and 1400 who has a better fit, and better research skills for that lab. With these two examples, they both have good numbers, but SOP and recs decide who will get into the lab. And if the whole applicant pool is full of people like them, then yes, the SOP and recs might be seen as more important, because they distinguish between them all.

[cranberry]

Knowing how to present yourself well is part of life. Knowing how to present research or an argument is part of academia. Plus, doing well in an interview can be the deciding factor in an acceptance or job offer.

Edited to add: I am as introverted and awkward as they come, so I know that reaching out and engaging can be difficult, but its necessary and I'm happy to do it.

Edited March 18, 2011 by cranberry

[Insei]

To the OP, science is undoubtedly about passion for research. But, without a good bit of creativity and the ability to work as part of a collective (e.g. interpersonal skills), you can't get very far. GPA and GRE scores are by no means an indication of those qualities. Science is a team sport; I think it makes a lot of sense that labs want people who will jive well with the group. I know that when I choose an advisor, how well we gel is going to be a huge determining factor. Spending up to 7 or 8 years working closely with someone is a big deal, especially if you hate each other

[truckbasket]

I'm in the humanities, so quantifying anything is problematic. With that said, I think there's an argument to be made that no matter what discipline you're going to work in, having strong communication skills (the rhetorical chops you suggest as necessary for the SOP), and strong people skills (the ability to develop connections you suggest as key for the LORs) is crucial to your ability to perform as a graduate student. Metrics alone cannot demonstrate your personality -- which must be a key factor if these programs are going to essentialy employ you for five to six years. It's the same in any workplace, hence the need for interviews and references in addition to the (relatively quantifiable) resume or CV. Just because you look good on paper doesn't exempt you from being a total psychopath.*

I could see there being a solid and frustrating argument for those who score well on standardized tests yet lack some of the social skills that can open doors; and likewise for those who cannot test for shit, but excel in all other areas. Clearly a balance of all aspects is what's needed, yet the GPA and the GRE are so full of variables that they just don't reflect the living, breathing human that the department is interested in meeting. By placing a greater emphasis on the candidate-as-a-numeric, you essentially strip away all of the candidate's character -- which is exactly what the committee is really looking for.

With all this in mind, and following an especially grim application season, I can help but sympathize with the monumental and daunting task that the admission committees have to deal with. It really can't be easy being in their shoes either.

* And here is where the SOP and the LOR work together, right? The LORs should confirm what was said in the SOP to prevent applicants from spinning off a bunch of made-up crap.

[wtncffts]

Yeah, I have to concur with the above replies in disagreeing with your points:

1) You're right to suggest that 'research ability', in a broad sense, should trump 'social skills', in a strict sense. Those 'objective' measures, however, can be very unreliable indicators. GRE, as we all know, doesn't test anything meaningful, or if it does, certainly not 'research ability'. GPA can vary heavily depending on the institution; even if it didn't, I don't think it's very indicative of research ability either. It shows your ability to do well within the pattern of undergrad courses, many of which aren't even in your field. Publications, I grant, pretty much directly evince research ability, but this also depends on the quality of the journal.

2) In terms of the SOP, as cranberry said, that "you have to know how to convince the reader, how to form the opinion you need through writing" is exactly what academia is. It has nothing to do with 'social skills', but skills of argumentation, reasoning, writing clearly and cogently, organization, etc. As others have said, "selling yourself" is also a crucial part of any job, whether it's a cover letter, statement of purpose, or interview.

3) I think you really overstate the LOR issue. You don't have to be best friends with your recommenders; you don't even have to be all that well acquainted personally. Granted, it might help; then again, it might cloud their judgment. All they have to know is enough to make educated guesses about your potential as a scholar. I hardly knew anything about my recommenders as people, and I certainly never was close to them socially. They knew my work, and drew conclusions from there. I think you also overstate the gullibility, as it were, of recommenders. Presumably, they're professionals: do you really think they're that easily swayed by your being nice, agreeing with them, and flattery? Now, undoubtedly this is sometimes the case; perhaps your experiences have led you to believe it. It has not been my experience, though, and I'd like to think it's not common.

Moreover, as someone else said, recommenders, if they're profs, are uniquely positioned to comment on your academic potential. They've been through grad school and are working scholars. This is the only part of the application which has that quality. Your SOP, writing sample, interview, etc. are all your own doing (frankly, there's no guarantee that an applicant's SOP and writing sample are even his own). As I said, GPA is a cumulative measure of how you did on very particular measures; tests, exams, assignments, etc. The whole isn't just the sum of its parts.

As everyone has said, social skills are important in academia and research, I dare say even more so in the sciences, where it seems your life is defined by labs and research teams, etc. It's always funny to me, as a non-scientist, to see the citation of a scientific article, with a dozen co-authors! In my field and in other humanities and social sciences, it's much more feasible that being a 'lone wolf', as it were, will get you where you want to go. In many areas, all you need to produce quality work is a brain, some books/sources, and something to write with/on. No social interaction required. Not so, I would say, in the sciences.

[ringo-ring]

Thanks to everyone for helping to clarify this issue.

I would like to add that communication skills, that are obviously critical for grad school success, are not enough for writing good SOP, and writing SOP is quite different from what you're going to do as a scientist - writing papers. In papers, what is needed is to prove a viewpoint on some issue in your field of interest, but in SOP, you have to prove your personal characteristics also - being passionate, dedicated, etc. - that's why it requires social skills too.

GRE test checks communication and argumentation skills isolated from all others, and it makes sure that it is student's own work, unlike e.g. writing samples. But even writing samples are a better alternative to the SOP here. But for some reason, they're not required for programs in sciences, only in humanities...

I think you really overstate the LOR issue. You don't have to be best friends with your recommenders; you don't even have to be all that well acquainted personally.

...

They knew my work, and drew conclusions from there.

But what graduate schools is looking to get out of LORs is not professional evaluation of your work, but again, social and personal characteristics. And for that some personal acquaintance is necessary.

Anyway, I agree that LOR is the best way to estimate social/personal skills (maybe not best, but I cannot conceive of a better alternative right now) BUT for any other skills such as research skills different methods of estimation should be employed, that treat applicant's abilities objectively, not just from the viewpoint of another person. Right now, unfortunately, LORs seem to be used as a substitute for everything, given that low influence other measures such as GREs have on admission process Even if they're written by professionals, these professionals are still people, and tend to view other people through the prism shaped by their personal experiences.

Clearly a balance of all aspects is what's needed, yet the GPA and the GRE are so full of variables that they just don't reflect the living, breathing human that the department is interested in meeting.

I agree that score on GRE tests and GPA are influenced by multitude of factors, but LORs and SOP are no less complex.

I do not argue for eliminating LORs and SOPs completely - they are very useful pieces of information. But they shouldn't be given as much influence and credit, compared to all other factors, as they have now. Instead, they should be treated equally or even less, since social skills, by which these measures are highly influenced, however important, is not the most important thing in doing science. All in all, they can be learnt/improved in the grad school already. Personal characteristics are also subject to change (of course if we do not talk about such important traits as dedication, but they should be evident from accomplishments not anything else)

Edited March 19, 2011 by ringo-ring

[truckbasket]

I would like to add that communication skills, that are obviously critical for grad school success, are not enough for writing good SOP, and writing SOP is quite different from what you're going to do as a scientist - writing papers. In papers, what is needed is to prove a viewpoint on some issue in your field of interest, but in SOP, you have to prove your personal characteristics also - being passionate, dedicated, etc. - that's why it requires social skills too.

I'd agree that there is a distinct difference between the humanities and the sciences; and consequently, the humanities people will most likely have somewhat of a leg up regarding SOP technique. However, I would argue that the SOP is not the place to prove anything at all. Proof, if anywhere, would be in the questionable metrics we've been discussing. Granted, the SOP is a tricky beast -- it's rhetorical to a degree, but must go to great lengths to disguise its own rhetorical methods. With that said, a great deal of research is required for the SOP regarding the fit aspect -- specifically aligning your potentially hypothetical aspirations with professors whose work you may only know in passing. It's less about proving your characteristics, and more about laying your assets out in a pragmatic manner that will give the committee the tools they need to make an informed choice. To me, it's more like offering a tasty buffet that entices through its inherent flavor, without relying on obnoxiously frilly shit like baroque linens and decorative silverware. But again, I'd refer to potential differences in disciplines that may shift the balance.

GRE test checks communication and argumentation skills isolated from all others, and it makes sure that it is student's own work, unlike e.g. writing samples. But even writing samples are a better alternative to the SOP here. But for some reason, they're not required for programs in sciences, only in humanities...

The GRE argument is a bit of a dead horse as this point, and there's plenty of other threads wrangling with it -- so I'll spare you my anti-GRE diatribes (I've done extensive research on standardized tests, although you've probably heard all the arguments). I would, however, point out that the communication and argumentation skills are problematic simply because many BA level writers actually have to devolve their writing skills in order to do well and please the GRE robot. The kind of writing that the GRE rewards is the kind of writing that many English profs spends years training their students not to do.

Anyway, I agree that LOR is the best way to estimate social/personal skills (maybe not best, but I cannot conceive of a better alternative right now) BUT for any other skills such as research skills different methods of estimation should be employed, that treat applicant's abilities objectively, not just from the viewpoint of another person. Right now, unfortunately, LORs seem to be used as a substitute for everything, given that low influence other measures such as GREs have on admission process Even if they're written by professionals, these professionals are still people, and tend to view other people through the prism shaped by their personal experiences.

Having been on a faculty search (the applicant's portfolio is very similar to what we've all been doing) I can tell you that the LORs didn't get as much love as the candidate's own letter -- simply because what we needed was the most concise and lucid portrait of the candidate available. Like you said, anything that comes from outside the candidate is filtered through a lens, and is probably more tinted (and varied) than the perspective the candidate can provide. The LORs served as evidential proof for anything referenced in the candidate's SOP. (Again, it's not an SOP, but you get the picture). In short, just like an interview or meeting someone in person, the SOP gave us the closest and most accurate way to judge the potential "fit" of an applicant -- far more so than their grad school grades, their class evaluations, and even the LORs.

I agree that score on GRE tests and GPA are influenced by multitude of factors, but LORs and SOP are no less complex.

Absolutely. But the difference, of course, is that the candidate has full control over the efficacy of the SOP. Other than studying your ass off, you basically have no control over what the ETS will throw at you (and even then, studying your ass off won't always work for a test such as this).

I do not argue for eliminating LORs and SOPs completely - they are very useful pieces of information. But they shouldn't be given as much influence and credit, compared to all other factors, as they have now. Instead, they should be treated equally or even less, since social skills, by which these measures are highly influenced, however important, is not the most important thing in doing science. All in all, they can be learnt/improved in the grad school already. Personal characteristics are also subject to change (of course if we do not talk about such important traits as dedication, but they should be evident from accomplishments not anything else)

Again, it might just be the difference in discipline, but I can't think of a single factor that presents you to the committee any better than your own words (other than the writing sample). To reiterate, the SOP allows you to present (depending on word limit, of course) everything that they need to know about you, without you jamming anything down their throats. It's allows you to allow them, to make the best informed decision about you that they can. And whether or not that garners an acceptance or a rejection becomes somewhat irrelevant. What I mean by this is that my own SOP was not designed to specifically win someone over, it was simply to present my skills and experience in a compelling manner. I truly believe that some of my rejections were not the result of my inability to convince anybody to accept me, but that I was able to provide a good overview of what I had to offer -- it just didn't align with their needs in the program. In other words, my buffet was well-presented, not garish, but it just so happened that they weren't hungry for the kind of food I was making.

Anyhow, good discussion!

Edited March 19, 2011 by truckbasket

[starmaker]

Research ability should probably trump everything - you're applying for research programs, after all - but what makes you think that GPA and GRE are reliable indicators or research ability? They don't and aren't meant to measure it. They don't correlate particularly well with success in research-based graduate programs. On the other hand, the point of letters of recommendation and the SOP is that you (SOP) and credible practitioners (recs) can talk about...wait for it...your research ability. Good SOPs and recs talk about concrete skills and accomplishments on the part of the applicant, rather than just indulging in flowery vagueness about how smart you are.

Publications are different. Peer-reviewed publications are actually very highly regarded by PhD admissions committees! After all, they demonstrate that established practitioners in your field reviewed your work and judged it up to the standards of a full-fledged practitioner - what stronger endorsement could you get for your ability to succeed in a research-based program? And even publications in places like undergrad journals are valuable, since they teach you about the process, which will be useful later on when you ARE trying to publish in peer-reviewed venues. But most applicants, even most good applicants, don't have any publications, so you can't just rank applicants based on number of publications (and not all publications are equal, either). Publications are more of an icing-on-the-cake deal - if you have them, it can help a lot, but if you don't have them nobody will look at you askance.

The problem is that there isn't a quantitative measurement of research ability that you can expect strong undergrads to have participated in. This is why we make do with flawed proxies - quantitative ones that aren't really about research ability, and non-quantitative ones that are.

If you really want to evaluate somebody's social skills, the best way to do that is probably the on-campus interview.

[Azazel]

As many others have posted, graduate school is principally about doing research. As test scores are poor predictors of research ability, schools don't tend to weight them very highly. I really think the issue is as simple as that ;-)

Unsurprisingly, the adcom's best proxy for your research ability is the opinion of a professor with whom you have worked closely (doing research!). So LOR are generally considered important.

[chaussettes]

As someone who probably would have benefitted immensely from a stronger weighing of "tangibles" (GPA, GRE) in the admissions process, I cannot in good conscience say that I think it would be a good move. Case in point, besides a high overall GPA and good GRE scores, I earned a high GPA in my Linguistics major, including some graduate coursework. I have a paper which I wrote for a grad seminar that I got an A on, which would probably be a fine writing sample to use. Yet I did not even consider applying to Linguistics programs because I realized that I do not really have the aptitude or motivation to do original research in any area of Linguistics. The only things that would bear this out would be my inability to formulate a half way convincing SOP and my (probable) failure to secure LORs from three professors who could speak enthusiastically to my research potential. I agree that this reliance upon SOP and LORs is problematic in some respects, but I firmly believe that, alongside the writing sample if required, they are by far the best indicators of research ability and potential.