An interesting glimpse into Rec Letters

[dugward]

Julie Schumacher helps expose the world of letter reading and writing in the Chronicle:

 

http://chronicle.com/blogs/conversation/2014/01/30/the-gristmill-of-praise/?cid=gs&utm_source=gs&utm_medium=en

[ToomuchLes]

I have supporting evidence for the second observation in the article. One of my strongest LOR was written by a retiring professor, famous for his contribution to African-American History of the 19th century. Before finalizing my LOR, I had the chance to peruse it, and I was shocked by the lack of a proper, professional format. Regardless, when I talked to a POI over the phone (casual non-formal conversation), I mentioned who I worked with, and the LOR I received; without missing a beat, the POI was like "ohhh I know that professor very well! Never met him but he is very well known in his discipline." 

 

I'd like to believe that LOR written by famous professors go a long way, but many forum users remonstrate the idea. 

[rising_star]

Schumacher makes an excellent point. My advisor has said he literally sends off 200+ recommendation letters each year, most of which likely go unread as noted in that article. If they decide based on your CV/cover letter that you don't fit with what they want, there's no way they're going to read all those rec letters. I'd say about half the academic jobs in my field are now only asking for the contact information for 3 references, which they only ask for letters for candidates who pass the first review. This makes soooo much more sense for everyone, though it may not be as practical for grad school applications as it is for academic job apps.

[TakeruK]

(I admit I had to look up what remonstrate meant).

 

I think the article makes a good point that not all LORs will be read. I don't think an admissions committee needs to review all 200 applicants for only 4 positions!

 

I also think that it's definitely true that the name of the LOR writer makes a big difference. I am not sure why this is viewed negatively by the GradCafe community (and I would say that I have not really seen much of this negativity [or I am misunderstanding what remonstrate means!]). What I see a lot instead on GradCafe is people arguing that getting a prof who doesn't know you well to write a letter just because they are famous would be not very helpful, and this is true. 

 

But, if the prof actually knows you well because you worked closely with them etc, then a letter from a well-known prof can be very strong. Not everyone's opinion is worth the same amount in real life and why should this be any different in academia? I think a strong recommendation from Nationally-Recognized Scientist would definitely be worth more than a recommendation from Random Prof No One Has Heard Of. But, it's also important that reputations of people can work both ways. Some people might be known to give overly strong endorsements and have students show up that aren't as good as their letters would indicate. If your letter comes from one of these people, then it will work against you. So, a well known name isn't always better! 

 

It makes sense for researchers to put more weight on opinions from experts that they trust or have positive experiences with. But it's almost impossible for an undergraduate picking which research position to take (and thus which LORs to receive) to know this. So while this is a fact, I think, there isn't much we can do to influence our future. Working for a famous professor in undergrad isn't necessarily going to increase our chances of getting a better LOR. They might be "famous" in a bad way to the places you're applying to, or they might be so famous and busy that they don't even interact with you and won't write you a strong LOR. 

 

So, this is why I would advise applicants to not worry too much about the famous-ness of their LOR writers. That's something we cannot really control. Instead, while working to earn these LORs, it's much better to do the best work possible (and thus find an advisor that can get you to do the best work possible). However, just because we can't control/shouldn't worry about LOR writer famousness doesn't mean that it doesn't play a role in evaluations at all levels.

[ToomuchLes]

@TakeruK

Im glad all the GRE words that I've learned, and continue learning are paying off lol. On a personal note, I've heard many times about the "Imposter Syndrome" that many grad students suffer, and so, I've taken the notion to continue making flashcards of words that I don't know the meaning to. Although, I realize that this particular ailment concerns the disciplinary knowledge, Im still very much afraid of graduate work surpassing anything I ever tackled.

 

Anyways, returning to the topic at hand, if not all LOR are read, then can we identity those specific LOR that are read?

I hypothesis that once an adcoms have a dozen or so potential applicants, they begin to really peruse the individual's info, along with his LOR; however, even a dozen applications possess 36+ LOR, and I truly doubt they'll read all. Therefore, since we (forum members) established that LORs are, in fact, an important staple, at what point do they actually become essential in the individual's acceptance?

[TakeruK]

Haha, I did vaguely remember seeing that word in my GRE list and I know I did learn it at one point (when I looked it up, the definition was familiar to me). But at this point, all I could remember (and deduce from context) was that it was some negative connotation, which would probably not be enough if this was a GRE question and I needed the precise definition to give an antonym or fill in a blank or whatever! 

 

Anyways, to throw in my opinion about "when LORs are necessary", I would say this: LORs don't need to be read to reject an applicant. Depending on their applicant pool and number of spaces, they might only skim LORs or use other means to narrow down their list. However, I think LORs are absolutely necessary to accept an applicant. I think it would be very rare if an applicant was accepted to a competitive program (i.e. pretty much all programs) without all parts of their package, including LORs, are reviewed. The author of the original article didn't say they read none of the MFA applicants' LORs--just that they only read the ones that had strong applications otherwise. They said that over 800 out of the 1644 letters went unread, which would mean this person only read the LORs of the top half of applicants (still a lot!). This very busy person still read about 800 LORs for their MFA applicants (4 spots!) so I don't think it's as extreme as 36 LORs being too much already. 

 

Also, I don't think the adcomms will cut it down so small to just a dozen shortlisted applicants if there are only 4 spots. In my field, I think profs try to make sure they get all sorts of applicants, not just those who have high GREs or GPAs, because sometimes the best students don't score super high in these metrics. So, I would think initial cuts are fairly conservative to avoid accidentally weeding out someone really good because of a low GRE score. However, there are still cuts for large applicant pools! I would think a single admission committee member is capable of fully evaluating much more than a dozen applicant. Assuming the author of the original article is a typical busy faculty member, this person read the LORs for about 100 applicants. Maybe reading LORs doesn't mean fully evaluating an application though, and maybe by "read" the author just meant "look at". Perhaps the right number of how many applicants a single admission committee member can fully evaluate is somewhere between 50 and 100? For large programs that admit 50 people at a time (and get over a thousand applications), I would guess that they split up the load between committee members. If I was doing it, I would arrange it so each application is viewed by at least 2 people but that is basically just a guess!

 

So, I don't think that this article means that LORs are not important nor does it mean that people can be accepted without their LORs read. The author definitely does not say that they did not read any of the LORs--in fact they said that they read about half of them, which sounds like a decent initial cut. Thus, my opinion is that the LORs may not always help you survive an initial round of cuts, but they will be very important to get you further in the process, and absolutely essential to getting an admit.

[12345678900987654321]

In my department's committee letters from people the committee members know carry a lot of weight. I cannot tell you how many times I've heard, "I know Professor Doe and trust her opinion." They'll sometimes call them up to get the "real" scoop or call someone in the student's department even if they didn't write a letter.

[VulpesZerda]

(I admit I had to look up what remonstrate meant).

I think the article makes a good point that not all LORs will be read. I don't think an admissions committee needs to review all 200 applicants for only 4 positions!

I also think that it's definitely true that the name of the LOR writer makes a big difference. I am not sure why this is viewed negatively by the GradCafe community (and I would say that I have not really seen much of this negativity [or I am misunderstanding what remonstrate means!]). What I see a lot instead on GradCafe is people arguing that getting a prof who doesn't know you well to write a letter just because they are famous would be not very helpful, and this is true.

But, if the prof actually knows you well because you worked closely with them etc, then a letter from a well-known prof can be very strong. Not everyone's opinion is worth the same amount in real life and why should this be any different in academia? I think a strong recommendation from Nationally-Recognized Scientist would definitely be worth more than a recommendation from Random Prof No One Has Heard Of. But, it's also important that reputations of people can work both ways. Some people might be known to give overly strong endorsements and have students show up that aren't as good as their letters would indicate. If your letter comes from one of these people, then it will work against you. So, a well known name isn't always better!

It makes sense for researchers to put more weight on opinions from experts that they trust or have positive experiences with. But it's almost impossible for an undergraduate picking which research position to take (and thus which LORs to receive) to know this. So while this is a fact, I think, there isn't much we can do to influence our future. Working for a famous professor in undergrad isn't necessarily going to increase our chances of getting a better LOR. They might be "famous" in a bad way to the places you're applying to, or they might be so famous and busy that they don't even interact with you and won't write you a strong LOR.

So, this is why I would advise applicants to not worry too much about the famous-ness of their LOR writers. That's something we cannot really control. Instead, while working to earn these LORs, it's much better to do the best work possible (and thus find an advisor that can get you to do the best work possible). However, just because we can't control/shouldn't worry about LOR writer famousness doesn't mean that it doesn't play a role in evaluations at all levels.

I don't want to hijack the thread at all, but I was relieved to read your post because I have been stressing about this issue. I think I made a thread about it a while ago but I'm still not sure after a recent conversation I had.

I worked for a lab (in my field) for two semesters. It was a fantastic experience and I'm still friendly with my supervisor and the PI. Then a lab (unrelated to my interests) that they often collaborate with asked if I would work for them the following semester. This didn't go quite as well. I mean, it was fine, but I didn't accomplish anything special, barely met the PI, and had issues about lots of expenses related to the assistantship coming out of my pocket (that they promised to reimbursed but never did). I was relieved when it ended.

Recently, when I was chatting with my supervisor at the first assistantship, she asked me who my third letter might come from. I said I'd love if she wrote it, since she has written one in the past, but she insisted the second lab's PI should do it since he is very famous. She knows about the issues I was having so I was surprised that she said that.

I am scared to ask him and can't imagine what he could even write. He *is* very famous, but I never thought that was a good enough reason. Plus, he's not in my field, if that matters. How can I tell my first supervisor that I don't think I want to listen to her advice?

[TakeruK]

I worked for a lab (in my field) for two semesters. It was a fantastic experience and I'm still friendly with my supervisor and the PI. Then a lab (unrelated to my interests) that they often collaborate with asked if I would work for them the following semester. This didn't go quite as well. I mean, it was fine, but I didn't accomplish anything special, barely met the PI, and had issues about lots of expenses related to the assistantship coming out of my pocket (that they promised to reimbursed but never did). I was relieved when it ended.

Recently, when I was chatting with my supervisor at the first assistantship, she asked me who my third letter might come from. I said I'd love if she wrote it, since she has written one in the past, but she insisted the second lab's PI should do it since he is very famous. She knows about the issues I was having so I was surprised that she said that.

I am scared to ask him and can't imagine what he could even write. He *is* very famous, but I never thought that was a good enough reason. Plus, he's not in my field, if that matters. How can I tell my first supervisor that I don't think I want to listen to her advice?

 

Are these two research experiences the only two research experiences you have? If so, the supervisors of these two projects should be letters #1 and #2, not vying for letter #3!

 

I think if you ended project #2 on good terms then you will probably still get a good LOR from them. It's better to have a research related LOR, even if you just did "nothing special" because research experience is still research experience. Not every undergraduate research position needs to turn out to be great science.

[VulpesZerda]

Are these two research experiences the only two research experiences you have? If so, the supervisors of these two projects should be letters #1 and #2, not vying for letter #3!

 

I think if you ended project #2 on good terms then you will probably still get a good LOR from them. It's better to have a research related LOR, even if you just did "nothing special" because research experience is still research experience. Not every undergraduate research position needs to turn out to be great science.

 

I'm thinking they would be possibilities for letter #3 under the assumption that my summer research goes well, which starts in about a week. I don't want to jump the gun since it hasn't started yet, but from what I understand, it is basically expected that I will ask my supervisor. It's one of those REU/SURF programs and I am a big fan of the professor!

 

And the other letter would be my major advisor, who is currently supervising my thesis...he also let me work with his grad students my sophomore year so he knows me the best and knows I've been interested in research for several years now.

 

So, if all goes well, all three letters will be from varying research experiences, which is why I'm being a little picky. I've been lucky to have many research experiences, but I rank them in my head and the lab with the "famous" PI wasn't great. It was only 5-10 hours per week and all I did was help collect data, nothing else. I didn't learn much.

[TakeruK]

Oh thanks for clarifying! That sounds good. I guess picking a letter #3 would be tricky then, because I probably would have done the same as you and gone to my first advisor first. But if she insists on you asking the other lab despite knowing the problems, then maybe she is trying to hint that her own letter won't be as good? Or maybe not, it's hard to tell! One potential solution could be to submit all four (not ideal but maybe a good idea if you can't decide between those two) or to submit one letter for some of the schools and another letter for other schools, depending on how well the letter writer is known at the schools you apply to.

[VulpesZerda]

submit one letter for some of the schools and another letter for other schools, depending on how well the letter writer is known at the schools you apply to.

 

^This sounds ideal! Thanks for all your ideas :-)

[joitomtom]

Insightful

[The Pedanticist]

Anyone see the last comment on the page?  It makes me shudder a bit...

 

 

When I was a young professor the Full Professor of Graduate Applicants

took the all 300 applications to his Lake Tahoe cabin and proceeded to go
on a month long bender. The Chair finally tracked him down and recovered
the applications - none of which had evidently been read and told me:

"Pick 30 - Only Ivy League or Stanford, Oxford, Cambridge and one German
and one French - rank them and have them back to me in two days."

God forbid...

[geographyrocks]

Just to confirm: who your letter writers are mean ALOT. My adviser told me today that grad students underestimate how important the letter writers title and position are.

[Sigaba]

IME, professors are so burdened with committee work on top of their regular responsibilities and their everyday lives that they're going to cut corners when they can. They recycle lectures and exams, they pre-sort applications they have to read, they have graduate students carry projectors to lecture halls. [Edit: I'm not bitter.] 

 

 

So, this is why I would advise applicants to not worry too much about the famous-ness of their LOR writers. That's something we cannot really control.

 

I respectfully disagree. One can do something about the prominence of one's LoR writers by applying successfully to institutions that have sterling reputations throughout one's pre-graduate school career. For better and worse, academic pedigree still matters.

 

 I think it would be very rare if an applicant was accepted to a competitive program (i.e. pretty much all programs) without all parts of their package, including LORs, are reviewed.

 

FWIW, I fell into that category. Part probably had to do with the length of my primary writing sample. However, based upon how some professors reacted upon hearing what school I attended as an undergraduate, my pedigree played a role. 

Edited August 13, 2014 by Sigaba

[Sigaba]

Just to confirm: who your letter writers are mean ALOT. My adviser told me today that grad students underestimate how important the letter writers title and position are.

 

By that same measure, if, as an undergraduate, you attend an institution where the lion's share of the teaching is done by graduate students, admissions committees will not automatically discount your LoRs if they're written by GSIs.

[TakeruK]

@Sigaba: Yes, you are right, we can and should consider the reputation of professors at schools we apply to so that we get LORs from well known professors. What I meant was that if you are already e.g. a senior undergrad applying to grad schools this fall, that train left a long time ago so there's no need to worry about things we can't change. But looking forward, choosing a graduate school with well known professors is important for LORs down the road, for sure!

 

Also, the longest item in any of my applications was a 2 page SOP. I was not thinking about fields with writing samples when I said that I'd expect the finalists at competitive programs would have all parts of their applications read.