Hacking the SOP.

[waddle]

waddle's statement-writing algorithm (adapted from the Feynman problem-solving algorithm):

1. Write down the prompt.
2. Think very hard.
3. Write down the response.

System requirements:

1. 2 weeks
2. your local library
3. a healthy dollop of copypasta from your NSF GRFP application, stirring/mixing well
4. LeechBlock, for denying yourself access to The GradCafe

Anyone care to post their method to guide this poor soul to The Way?

Edited January 4, 2011 by waddle

[katerific]

I'm pretty sure my initial drafts started with a stream-of-consciousness sort of style. I even doodled things, like thought clouds and such. Even though I thought I'd be able to clean it up and forge it into a coherent piece, my work could never make me happy, plunging me into deep despair and an existential crisis. After drinking heavily and doing some abstract expressionist painting (with many cigarette butts incorporated into the canvas), I returned to my work. After a while, I found the spark that allowed me to create a beautiful(????) piece that ended up being submitted to multiple institutions. (Well, I did a set ("Variations on a theme") and made individual pieces for each institution, of course.) Direct, cogent, smooth (yet raw), this baby makes a STATEMENT.

Okay, it wasn't quite that dramatic, but it was something like that.

Note: I don't recommend that people do this. I mean, unless that's how you roll, then I guess it's okay.

But in all seriousness, the best thing I did was give myself enough time to ruminate. Like, put it down, let the dust settle, and pick it back up.

[Soheyl]

I used a concept-mapping software to manage and organize my thoughts about (1) what to say that addresses adcom's main concerns about an applicant and (2) how to say them in a way that reflects the qualities I want the adcom to see in me. This was the toughest part since I wanted to say something that other applicants don't say, not because they don't want to; but because they can't. I'm not sure if I was successful in this, but I did my best.

Edited January 4, 2011 by Soheyl

[UnlikelyGrad]

I've written not just an SOP, but a bunch of personal essays for fellowships (2 years in a row). Here's what I do:

(1) Look at the prompt.

(2) Brainstorm what character attributes I have that will impress people who want to hear about the prompt.

(3) Consider anecdotes/data/experiences/etc. that prove that I have those attributes

(4) Write up paragraphs involving stuff from #3

(5) Put paragraphs in order and add transitions

(6) Think of a really good 'hook' for the beginning

(7) At this point the draft is almost always way too long, so I go through and examine which things can be shortened, and which things are unnecessary enough that they can be cut completely.

(8) Now I write a conclusion. Some people might do this earlier in the process but I have a hard time writing conclusions.

[BlueRose]

My algorithm looks a lot like UnlikelyGrad's, only with more cursing.

For fellowship and first-choice school:

1) Write down the prompt.

2) Carry prompt around in pocket for a few days. Think about stuff that I want to say. Write little notes to self.

3) Read statements that other people have written. Take mental note of elements I can use: structures, formatting, rhetoric. Also note what doesn't work and why.

4) Organize notes into vague outline of talking points. Edit until satisfied with content, logic, and projected length.

5) Write crappy first draft. I have to force myself to not see it as a submittable, judgement-worthy artifact. So I write by hand, in pen (no delete key), in my worst handwriting, with liberal use of LOLspeak and swear words.

6) Repeat #5 as necessary.

7) Type up lunatic scrawlings from #5 and #6 and edit into a presentable rough version.

8) Starting with the most important bits, go into polishing mode. I start with the important bits because the polished version is usually longer, so something might have to go...also I might run out of time.

9) Read final version. Decide that some paragraphs are in the wrong place. Move them. Inevitably, I've written a decent conclusion somewhere in the middle. Put that at the end.

For all schools:

10) Re-read department website, taking notes.

11) For candidate advisors, read / re-read at least a few key papers.

12) Edit "fit" portion of essay, weaving in bits about advisors, resources, etc.

For all schools after the first:

13) Write down prompt.

14) Realize that the school wants the same information, only distributed over a different set of essays. Sigh deeply.

15) Bribe self with Diet Coke. Let the copy-pasting begin!

16) Read resulting essays to make sure they still make sense.

17) Read them again to make sure all references to first-choice school have been deleted.

For all:

18) Submit!

19) Wake up in the middle of the night and realize you've said something stupid / untrue / nonsensical.

20) Curse self. Wonder if McDonald's is hiring. Wonder why they would hire you.

21) Re-read essays. Yes, you really did say that. Curse self again. But hey, the rest of this isn't bad...

[UnlikelyGrad]

My algorithm looks a lot like UnlikelyGrad's, only with more cursing.

Hey, just because I don't intentionally write cursing into my algorithm doesn't stop it from happening at regular intervals.

[lab ratta-tat-tat]

I could be totally wrong but here it goes...

My boss (who is on the admissions committee at University of Awesomenes) told me it is always important in your first paragraph to state your purpose. Why are you applying to graduate school? What is your purpose? A lot of people do not state that and it drives adcoms crazy. Your first paragraph can be a brief introduction about how you got interested in your field, some experience you have in it, and lastly, what motivated you to seek a PhD? Was it the experience you had as a research assistance was so great? Was it the little career advancement you saw in filed X as a basic researcher? Whatever it is that motivated you state it, whether it is at the end of the paragraph or the beginning, adcoms want to know why are you applying, and why should they invest $25,000 in you for the next 5+ years and are you worth it? If so, prove it, show them.

Also, it may seem really silly but if you have a writing center on campus, use it. They have people who specialize in technical writing, scientific writing, admission to graduate school writing, grant writing, etc. Also, the writing center is free. I went there 3 times and at my school, you were able to email them your drafts if you could not meet with them.

Something interesting, I was told was to not write a lot of crap about every research position you ever held in the past. It is great you were involved in lab Totally Rad for 1 year and lab Kinda-not-so-awesome for 3 years, but you don't need to list every thing you did, they can look at your CV to see all the positions you held. Highlight the most important ones that speak to your potential future as a graduate student in their program.

Lastly, it is important to state your long term career goal(s). This can be vague such as "a career in academia" or whatever. Not everyone knows exactly what they want to do, and not everyone wants to be Principal Investigator (PI) and that is fine. Point is, they want to know you can complete their program and go out and get a job and represent their program well.

This is all advice I received from students and faculty members on adcoms. It is not applicable for every school or program but just a overarching outline of what to address in your SOP.

I hope that helps and does not come off abrasive.

[BlueRose]

...it is always important in your first paragraph to state your purpose.

Agreed. I tried to go for the "hook" approach, but it just felt forced. So I cut the cutesy childhood story and got right down to business.

...if you have a writing center on campus, use it.

Also agreed. I did that my first year. You may have to ask around for the kind of help you want; some people will work with you on big-picture stuff, others are better with the nit-picky grammar, so decide what you need and say so when you're scheduling an appointment.

I was told was to not write a lot of crap about every research position you ever held in the past.

I wish more schools accepted CVs. I took this approach whenever I could get away with it. Otherwise, they got a shortened version of my NSF essay tacked onto the end (a prose CV, more or less). I did highlight the relevant parts in the "fit" section, but I would have done that anyway. I'm working on my last school now, and they don't even take supplements, so I'm going to have to add the teaching portion of my NSF essay too. Eventually. In a not unrelated note, I've been posting a lot today.

Edited January 5, 2011 by BlueRose

[lab ratta-tat-tat]

I also have been posting a lot today. I think it is to avoid the feelings of angst, anxiety, and an emotional breakdown. Any free time I have to consciously think, I may break down and cry.

But seriously, I saw this topic and figured I would share my experience in writing SOP's since I am happen to know people on adcoms who are willing to give great free advice, i figured I could give some insight on the issue.

[katerific]

I also have been posting a lot today. I think it is to avoid the feelings of angst, anxiety, and an emotional breakdown. Any free time I have to consciously think, I may break down and cry.

But seriously, I saw this topic and figured I would share my experience in writing SOP's since I am happen to know people on adcoms who are willing to give great free advice, i figured I could give some insight on the issue.

You are amongst friends, friend! And I think everyone appreciates a different perspective on things. I mean, not everyone may agree on the way to write an SOP, but it's certainly good to keep all these things in mind, at least. And I have to agree with most of what you said; I'd probably add that when highlighting certain projects, it's good to be specific yet succinct to demonstrate your ability to communicate your knowledge of the project/subject. (Quite a few profs stressed this point to me.)

This thread is taking an interesting direction (and now I feel silly for posting such a lame/silly thing! Haha.).

[UnlikelyGrad]

I cut the cutesy childhood story and got right down to business.

A hook doesn't have to be a childhood story. It can easily be stating your purpose.

I wish more schools accepted CVs. I took this approach whenever I could get away with it.

I just sent them a copy of my resume whether it was asked for or not.