Some Suggestions on writing your statement of purpose


From: thane 

	Thanks for all of the quick responses!  I've included the
following in my compilation:
	1. original query
	2. SOPs
	3. Remaining comments 

	Also, I have recieved a copy of the original SOP compilation.
If anyone else is interested in it, I'll be happy to forward it to you!

1:
=========================== ORIGINAL QUERY =============================
Systers--

I recently requested a past compilation on "statement of purpose"
essays.  While many systers have contacted me also desiring a copy,
no one has come forward with one.  So, I am volunteering to make
a new compilation.

Right now, I'm in the process of applying to graduate schools, and I'm
really at a loss as to what I should discuss in my statement of purpose.
I don't feel that I've been exposed to enough CS to declare what I want
to study. . .everything I've learned to this point has been fascinating.

So, I guess I want to know what is appropriate to write?  Is it our
time to shine (ie: brag), or simply state our current interests?  This
is of particular interest to me since I think it might be harder for a
woman to "brag" than a man (not that I want to make sweeping
generalizations, but I know that I "brag" to my professors *a lot* less
than my male counterparts).  Basically, what is the purpose of such
an essay?  I'm sure there are many of us that would greatly appreciate 
any advice/ramblings :-)

Please send all responses to me only.  I shall strip all names and
designate as forwardable unless otherwise stated.

Kay Hane
thane@crow.ucs.indiana.edu
==========================================================================
2:
=============================== SOPs ===============================

Here you go...  FYI I got into MIT, CMU, and Berkeley with basically
this essay, and was wait-listed at Stanford.  Of course, it's hard to
tell how much of that is from the essay and how much from other parts of
the application.

I ended up in an area pretty close to that described in my statement of
purpose, but I wouldn't say that that was necessarily true of statements
of purpose in general.

- -------------------------------------------------------------------------------
The question I want to study is as follows : how does the
implementation of intelligence, or perhaps more aptly the organization
of knowledge, delineate a being's view on the world and itself?
'Being' here is meant quite loosely, as any system that uses world
information in its interactions with its environment. It can apply
to anything from a tadpole to a human to a natural language
understanding system.  While this question itself is quite vague, its
meaning becomes clearer when taken in the context of a particular
field of study. In linguistics, it implies questions about the
relationship between language and thought, and how the structure of
language affects one's view of the world.  In philosophy of science,
it asks to what extent our laws of nature can be considered objective,
and to what extent they are merely manifestations of the way we
organize information about the world.  In artificial intelligence, 
it asks whether we have an objective enough view of our own
intelligence to build another intelligent system, and how our
implementation of intelligence might determine or limit the system's
interaction with its environment.  Because these questions of
intelligence have ramifications in many fields, an appropriate angle
of attack for this problem is an interdisciplinary one, bringing
together the tools of computer science, philosophy, cognitive
psychology, and computational linguistics. 

My interest in this problem has sprung from a number of sources.  My
original interest in computer science was sparked by work at the Naval
Research Laboratory Connection Machine Facility, which provided me not
only with my first taste of the joys of parallel computation, but also
with exposure to the wide range of work currently done in computer
science.  I am free there to dabble in whatever interests me, and my
work in numerical analysis, artificial intelligence, and computer
graphics have given me a better feel for what research in computer
science is really like. Since then, courses in artificial intelligence
and machine learning have fostered an interest in the relationship of 
artificial systems to the world, and whether it is possible for one
intelligent being with great gaps in its understanding of its own
intelligence to build another one.  Automata theory, especially the
work of John Von Neumann and John Holland on self-reproducing
automata, has provided a rich source for musings on the extent to
which automata can "know" about themselves and their environment.
Interest in problems of language, knowledge representation, and the
subjectivity of humans as observers in the world has been stimulated
by my work in Germanistics through contact with the work of German and
Austrian philosphers, most notably Nietzsche and Wittgenstein.  The
scope of the (in)ability of people to understand their own cognitive
processes has interested me since studying readings of Freud. 

I believe Carnegie Mellon is a good place to study this problem
because of its great strengths in Artificial Intelligence, in Computer
Science in general, and in Computational Linguistics, in combination
with the University of Pittsburgh.  An integrated program with a
philosophy department is a good sign that questions such as those I
would like to pose may be answered within the framework of the
graduate program in Computer Science, which may not be the case at
some of the other big-name schools. If I come to Carnegie Mellon, I
want to get a thorough grounding in computer science, not only in
artificial intelligence, but also in areas such as automata theory and
robotics, which provide alternative approaches to the problems of
autonomous systems and self-reference.  A strong background in
computer science will give me a solid basis from which to work in
exploring the problems that interest me.  In addition, I want to
continue moving in an interdisciplinary direction by studying related
fields of the humanities.  In particular, I want to develop a firm
foundation in philosophy.  In short, I want to learn to combine the
analytical method of thinking that characterize science and
mathematics on one hand and the analogical method of thinking of the
humanities on the other into an approach that will be appropriate for
studying the problems of intelligence and knowledge representation in
cognitive systems.  

==========================================================================

Throughout my undergraduate career, I have tried to obtain a wide
variety of experiences so my graduate studies can build upon this
solid foundation.  I have worked in industry, in a professor's
research lab, and on my own individual research project.  I have also
gained valuable leadership and communication skills by participating
in many extra-curricular activities.  My diverse background and strong
academic performance prepare me well for graduate school.

As an engineering intern at Honeywell for three consecutive summers, I
learned how a product is designed, tested, re-tested, and brought to
the market.  I also gained experience with automation and control by
writing the software to automate the efficiency test for electronic
air cleaners.

As an aide to a grauate student in a physics research lab at the
Univeristy of Minnesota, I wrote data analysis software and helped
with the testing for an experiment which analyzes the gaseous
composition of cometary dust.

This academic year, I am doing a senior honors project in EE.  My
research involves the design of a temperature sensing system which
will monitor and record breast surface temperature over time.
Evidence suggests that abnormal rhythmic variations in surface
temperature may indicate that the underlying tissue is cancerous;
therefore this system may become a pre-screening for breast cancer.

I have also been involved in many student organizations, and I served
as president of three different groups.  Through these experiences I
learned to work well with groups of people and to co-ordinate their
efforts to reach a common goal.

At the University of California at Berkeley, I want to study
solid-state devices and semiconductors.  During my first year there, I
will take a number of courses relating to the characteristics of
solid-state devices.  I plan to learn as much as I can about their
operation, and especially the factors that limit their performance.  I
will study the semiconductor materials that are used to make these
devices, and develop an understanding of the physics of their
operation.  I hope to also be involved with a professor doing research
in this area, so I will get some experience in fabricating and testing
these devices.

After I have acquired enough educational background, I want to do some
independent research into ways of improving the performance of
solid-state devices.  In an effort to optimize the performance, I will
experiment with varying one parameter and then measuring the resulting
device characteristics.  Through repeated experimentation, I hope to
be able to fabricate an improved device and theoretically explain its
improvement.

 After I complete my M.S. and Ph.D. degrees, I want to pursue an
academic career involving teaching and research at a university.

==========================================================================
3:
=========================== REMAINING COMMENTS ===========================
Hi, Kay:

The "statement of purpose" is a very important piece of information along with
your GRE score and your undergradue GPA when you apply graduate school.
It is even more important if you are applying for a tuition aid.

There must be a reason or reasons why you want to pursuit graduate study, 
why you choose that particular graduate school.  Tell them "why", then
"what" you plan to study and "how" you plan to achieve you goal and objectives. 
I am sure you must have asked these questions to yourself many times, now it is
time for you to tell the school board members who select graduate students.

> So, I guess I want to know what is appropriate to write?  Is it our
> time to shine (ie: brag), or simply state our current interests?  This
> is of particular interest to me since I think it might be harder for a
> woman to "brag" than a man (not that I want to make sweeping
> generalizations, but I know that I "brag" to my professors *a lot* less
> than my male counterparts).  

"Bragging" is a personal style.  Some people feel comfortable and some don't.
Someone may be get selected to graduate school by bragging, but whether he/she
can complete his/her graduate work is another story.  I still remember there was 
a guy sponsored by Bell Lab in my class when I was at the graduate school.  
He is "Mr. B.S.".  I did the first project with him and the other woman. 
After two weeks, we told him that we do not want him as our partner.  
He B.S. or bragged about anything he wanted but no one was willing to do projects 
with he any more.

Best wishes to your graduate study!

===========================================================================

I've been told they want you to state a specific topic you would be
interested in, even if you really have no idea.  You can change your
mind later in most schools without much trouble.  Some schools may
use this to assign you an advisor, though, so be careful to do it in
such a way that this will likely work out for you if you are applying
to a school that assigns advisors before you get there.

==========================================================================

My impression is that the statement of purpose isn't very important in
the decision to admit.  When I participated in admissions I read all
of the statements and some of the others on the committee also did
this, but many committee members ignored it all together.  At least
for this school (in the "top five") what counted most were the
quantifiable things: grades, letters, and GREs.  The statements are
useful for two purposes: explaining any anomolies in your record and
helping the committee determine what area(s) you are interested in
order to gauge how many students in each area will be entering the
department.  Of course most forms require you to state your area
somewhere other than the letter, but this is a useful place to mention
the names of faculty members with whom you would like to work, if you
are familiar with what work they do.  This strategy is useful only if
you've had some kind of contact with the faculty member, but can
backfire if that faculty member doesn't know you and isn't impressed
by your record.

Other than these two things, I would be sure to state any kind of
research experience you have had, and put it in its best light.  The
committee wants to accept people who can show they have strong
potential for excelling in research.  Of course this advice doesn't
apply if you are trying to be admitted for a master's degree, as
opposed to a PhD.

=======================================================================

Kay -

I'm no expert, but when I did my statement of purpose I assumed the
purpose was for the dept to which I was applying to see whether
there was a good match between my interests and the interests of
someone on the faculty there.  I just described the kinds of questions
I found interesting and issues I wanted to study, all in pretty
general terms, and I was pretty upfront about the fact that I
wanted to explore several areas before settling on one.  I hope
some folks who have been on the other end of the process will
answer and say what they looked for.

(By the way, I did get into a top-3 dept, so I think my statement
wasn't completely wrong-headed.)

===========================================================================

Kay,

In my opinion, a statement of purpose is where you tell the professors why
you want to go to graduate school.  When I read applications, I want to
know *why* this person wants to go through the pain of several more years
of very difficult classes, a lot of "hoops", late late nights for no pay,
etc.  Why do you want that unpleasantness for the next 6-8 years of your
life when you could be out making $40K a year for a 40-hour work week -
what is it that *you* want to do in the world that you can't do without a
PhD?  What makes you think you want to do whatever-it-is and what makes you
think you can't do it without a PhD?  

BTW, this doesn't apply for a masters - a masters is short, not that much
different from an undergraduate degree (there not as painful) and gets you
more money when you get out - so the motivations are pretty obvious.  PhD
programs on the other hand are much more difficult, much longer, and mostly
degrading and demoralizing unless you have a "vision", a "burning desire"
that will see you through. The statement of purpose is the place to discuss
that burning desire.


You can use my name.

Bonnie

Bonnie E. John
Assistant Professor
Computer Science and Psychology Departments
Human-Computer Interaction Institute
School of Computer Science
Carnegie Mellon University
Pittsburgh PA 15213
Phone:  412-268-7182
FAX:    412-681-5739
E-mail: bej@cs.cmu.edu

==========================================================================

The essay should indicate what you want to study and why.  It is used
to determine if the applicant's interests have any relevance to what
the department can offer, if the applicant seems to have any true
interest in the subject, and the depth of the interest.  Bad statements
convey the attitude of "graduate school is what comes after 16th grade",
or "I didn't like physics so I decided to try something else".  Good
statements do not brag (tangible accomplishments should be documented
as grades, awards, etc. in other parts of the application), but they do
relate experience with desire, e.g. "liked theory courses, especially
interested in graph theory, have experimented with traveling salesman
algorithms."  The statement should have some indication of how graduate
study fits into post-graduate plans (something more than "it would be
good to have a graduate degree.")

Having said all that, I'll go on to venture the opinion that the essay
isn't very important.  Most statements are so very bland and careful
that they are disregarded.

========================================================================

Kay,
I'm so sorry nobody came forward with a copy of the compilation!  Perhaps 
someone would offer her copy after your plea.  I know the sort of anxiety
one has when writing a statement of purpose, I had been there myself.  
Some books in the library/book store can be useful in pointing out what 
sort of things faculty and school officials are looking for in an applicant.
I remember one book says that while writing an UNDERgrad essay the emphasis is 
to appear MATURE, in writing a GRADuate essay, on the other hand, the emphasis
is to appear PROFESSIONAL.  You can certainly talk about what interests you
and what you've acomplished in your college career, but more importantly you
have to project yourself as one who's eager to pursue graduate studies in the
field: you have to show that you love CS and you know what you want to do in 
grad school.  I know that last point must sound unreasonable (how do you know
what you can do in grad school?), but as my advisor at MIT used to say, "We
know regardless of what students put on their applications they're going to 
change their mind some time along the way, we never expect them to really do
what they say they'd do on their essays.  It's just that we'd like to hear 
what sort of interests and professional aspirations they've, and what they
think of their work and CS in general."  Something like that.  You know,
faculties just want to see whether the applicant "has a clue" or not.

Also make sure not just the content but the presentation is equally
professional: I'd definitely get it printed on a nice laser printer, 10-12 pts
Times font + 1-1.5 spacing would look good.

Good luck!

==============================================================================

Kay,

When I applied to grad school almost 6 years ago, the story I heard
was that it didn't really matter what you said in your essay, as
long as you had an idea of what you wanted to do.  I had no idea what
I wanted to study, so I picked a topic that sounded interesting and
wrote about that.  I also tried to talk about my past experiences, 
including internships and a small research project at school, and why
I thought that they would help me in grad school.  I also talked about
why I wanted to go to grad school, and I think I even mentioned my
extra-curricular activities (I was involved many student organizations)
saying that I had learned how to work with other people very well 
(alot of research is done collaboratively).  If I can find my old essay, I 
will send it along.  I think it's on a Mac disk somewhere.  It must
have been okay, since I got accepted at the top three schools (in EE).
I did not end up studying anything even remotely related to what I
wrote about in my essay.

Now I am almost done with my Ph.D., and this year I sat on the admissions
committee at Berkeley.  I read about a hundred applications, and I think
that the advice that I had gotten was actually not too bad.  The way
that the process works is that the applications are sent out to the professors 
to read, which professor reads your app. is chosen by the area of interest 
stated in your application.  The most important part of the application is
the letters of recommendation.  If another professor at a good school thinks
that the student will make a good Ph.D., that goes much farther than GRE
scores or even grades (although both are important).  Most people don't have
such letters, and so grades and scores factor in more heavily.  From
what I saw, the essays were mostly a +/- sort of thing.  A really good essay
might get you a +, a really bad one will work against you, but most of them
are middle of the road and don't count for all that much.  If your essay 
is really specific about exactly what you want to do, that could work 
either for or against.  If one of the professors gets excited about 
your ideas, you have a much better chance to get in.  (If you really have
such great ideas, you might be better off telling the professor you
think will be most interested over email.  Don't depend on luck for the
right person to read your application.)  However, if you say you really 
want to study a very specific problem in area X and the department 
feels that they have no faculty in that area, or the one person in that 
area is on sabbatical, or not taking any new students, or very short on 
funds, or that person is just not interested in your specific problem 
(and thus is not excited about having you as a student),  then you will 
have a worse chance of getting in.

What you might not know is that admissions are done by area (at least here
at Berkeley).  The department can admit N students.  Area 1 gets n1 slots,
area 2 gets n2 slots, etc, and each area decides which students to admit.
Since there are some areas with lots of students applying, it is very hard
to get into those areas.  Some areas have very few students applying, and 
thus it may be "easier" to get in.  It really works against you though, if
you apply to area 2, thinking that it will get you in, and then try to
switch to area 1, which is more popular.  In the most popular areas, it can
be really hard to find an adviser, get funding, etc.  That is why they do
the admissions by area, they actually admit only as many students in each
area as they feel they can support (both financially and with advisers).
A few always end up switching, but they do try to keep it balanced.  But if
you honestly don't know what you want to study, and you can find out what
areas are easier to get into at the school you really want to go to, you
might want to consider this.

My advice to you, if you don't really know yet what you want to do, is
to choose a topic that sounds interesting (nothing so narrow as a Ph.D.
topic, but an area that is well-defined enough) and learn something about
it, try to read a couple of papers, or better yet talk to people at your
school who are working in that area.  Then write a general essay about why
you think the topic would be interesting to study, and what skills/experiences
in your background/schoolwork would help you do research in that field.
It probably helps if you know enough about what is going on at the schools
you are applying to to be able to mention that in your essay.  Most people
don't do this at all.  I don't think you don't need to "brag" in the sense 
that you just detail all your accomplishments and assert that you're going
to be the hottest new grad this year.  Definately you should sound intelligent,
interested in or even fascinated by the topic you have chosen, and also
confident.  Like I said before, you should talk about how your past experiences
will help you in grad school.  Try not to sound wishy-washy, or like you are
going to grad school because you don't know what else to do, and you have
no idea what you want to study.  

==========================================================================

my statement of purpose was quite definite in what i
wanted to do, and while that apparently was appealing
to some (ie, berkeley offered me a 3-year graduate
fellowship for the phd program in cs, and i got
an nsf 3-year graduate research fellowship),
to others, i was 'too focussed' (hertz fellowship
interviewer).  my impression is that being
very focussed will make you appealing to those
who agree with your focus, and unappealing to
those who disagree with your focus.  if you
say it all fascinates you, and have great
grades and letters and gre scores, then i
can't imagine anyone being unhappy.  they'll
probably try to recruit you for their favorite
field/research project.

========================================================================

Kay,

Thanks for offering to make this compilation.  I look forward to reading it.

A few comments of my own:

I've heard it said that the essay in an application matters very little.
I have a good friend on the admissions committee who says that decisions
are made on 1) recommendations 2) courses and grades 3) GRE's
and 4) the essay, in that order.  The essay, apparently, is used more to 
screen out the loonies ("If I don't graduate in 3 years I will bash my
advisor's head with a hammer!") than to measure the candidate's skills.

That said, I still think it's worth making your essay as good as possible.
My friend could be deluded or my school could be unusual.  Here are a
few thoughts --

1)  often a "less qualified" student will be admitted if his/her interest
closely matches that of a faculty member that wants students.  So, even
if in truth your interests are broad, write down something specific that
you think will attract the attention of particular faculty.  For example,
at my school we have several database faculty who want students.  But 
almost no applicants ever indicate interest in databases.  So if somebody
writes "I'd like to do research in the area of databases," they are more 
likely to be admitted, regardless of GPA, etc.  Saying something in your
essay does *not* commit you to doing research in that area, but it does
get your foot in the door.

2) when you talk about your previous work in the essay, make clear what
you did independently.  Lay out clearly what hard problems you solved.
Give some indication that the work was successful -- users liked it,
it integrated seamlessly with the overall system, you published a 
paper, or something.

3) in general, go ahead and brag.  But remember, it's more powerful 
to give specifics than broad generalizations.  For example, "I am
highly motivated and a smart person," is less strong than "On my
own initiative I implemented a such-and-so which turned out to have
the following beneficial properties..."

4) I used a fairly informal tone in my SOP essay.  Remember, the
committee reads hundreds of stodgy boring essays.  Tasteful humor
in moderation can help them remember who you are.

OK, I'm running out of steam.  Good luck!

===========================================================================

	I would suggest you try to be as specific as you can about something
that excites you in your statement of purpose.  Being specific, gives them
something concrete to evaluate you on.  I did something along the lines of
the following:

	My interest in CS began when I found that the most interesting
	discussions in a linguistics course (on syntax and semantics)
	were not those among linguistics students but rather the CS grad
	students in AI. I dicovered that computers can generate examples
	and test theories much more effectively than can humans, making
	weaknesses obvious much more quickly than is possible using traditional
	methods.  I began taking CS courses to equip myself to use this tool
	and over time, my fascination with computers continues to increase.

	Simulation and modeling are analytic methods that are appropriate
	to many types of problems.  I'd like to develop these areas, using
	computers.  In the long term, I would like to either continue developing 
	tools or use my understanding of the tools by applying them to set up
	and solve real-world problems.

=============================================================================

I'm just a graduate student, but my advisor has served on and chaired
our admissions committee and he has been very open about why they ask
for a statement of purpose and what they look for in reading them.
According to him, the number one purpose of this essay is to determine
(as much as possible) that department and potential student will form a
good match.

There are two main issues here.  First, no school does everything.  It
is extremely unlikely that you will be able to complete a dissertation
in a topic that is not supported by at least one professor in that
department.  For example, I think that my institution is a pretty good
one - but we don't do robotics at all.  If you think that robotics is
an area that you will likely want to explore, you should be applying to
a school that does robotics, because you simply will not be able to do
that here.

Assuming that the department does the things that you want to do, the
second major issue is whether the professor(s) that you are likely to
want to work with are accepting students at this time.  I once saw a
student travel halfway around the world in order to work with a
particular professor - only to discover after he arrived that the
professor in question was not accepting new students.

Finally, I don't believe that anyone expects you to know exactly what
you want to do.  But they do expect you to have done some basic
homework about the department and to be prepared to explain how your
probable interest areas mesh with those of the department.  To do this,
you need to know what the various faculty members' research interests
are.  If at all possible, you should contact potential advisors in
person or by email and discuss potential research areas and whether
they are accepting new students.  No, they won't commit to accepting
you as their student sight unseen, but at least you will find out if
there's any hope and get some indication of what they will be like to
work with.

One admissions committee member - not my advisor - commented that the
kind of application he likes to see says "I look forward to studying X
with Dr. A."  If you can't narrow down your interests that precisely
yet, at least talk with several potential advisors and write an essay
that says "I have narrowed my interests to X, Y, and Z and have
discussed these with Drs. A, B, C, and D.  I look forward to exploring
these interests more fully at your institution."

They also look for secondary things like a basic ability to express
yourself in writing (crucial for getting published) and some indication
of who you are.  By all means, brag as much as you can force
yourself to:  "During my internship at Prestigious Lab, I found myself
particularly fascinated by our work in Topic X."  But I really believe
that the most important thing is to show them that you know what their 
department does and that your probable research interests and theirs are
a good match.

=========================================================================

I applied to graduate school a year and a half ago.  I was
really pleased with my statement of purpose.  It included
sentences like these:
  "In my six years of experience at Portland State University
   I've been the picture of an alternative student."

  "The ratio of men to women was similiarly low at the Graduate
   Theological Union where I studied in 1982 and 1983.  The
   female students were, as a group, noticably more able than
   the males.  I attribute this discrepancy to the unwillingness
   of mediocre female students to persist in Theological
   studies.  I look forward to the day when the women are as
   mediocre as the men!"

  "As a graduate student of CS at PSU, I will be pleased to 
   contribute to the numbers of women in computer science -
   though not their mediocrity."

I was pleased with this strain because it
	1. reflected my convictions and beliefs
 	2. involved ideas about education
	3. was clever
	4. capitolized on the fact that the people in the
	   department had known me as a student for
	   years.

My advice:  write about your ideas about your life (if you are
comfortable with that sort of thing).  Paint a picture of yourself
as a scholar and thinker.  Mention areas of interest.  Keep in
mind the audience consists of people who are enthralled with the
research and education process.

======================================================================

Write about your accomplishments to date.  Write about projects
you've worked on.  Write about what interests you about the projects,
about computing and about that school.  A common approach to
graduate school is to decide what area of graduate study you are
interested in and to find out what schools specialize in that area.
Often you find out who is writing in that area and apply to their
schools.  Your application should reflect this.  The school can
then see if your interests are a good match for theirs.  The
school will look at other factors as well, such as GRE's and GPA.\
Basically this is your chance to attract the attention of the 
people that you have already decided you want to work with.  Tell
them what you are excited about and why.

======================================================================

I have been on the admissions committee at a large "top-ten" (in CS)
university for 2 years. Our committee basically disregards the statement of
purpose, so it doesn't much matter what you write. Think of it as a writing
sample (make sure your English is good --- that will be noticed), and talk
about what you like about CS, and why.

The only thing you don't want to do is proclaim that all you've ever wanted
to do is X, when the dept doesn't have anyone doing research in X. X might
be computer-human interaction, or databases, or performance analysis ---
all good areas of research, but not at all schools. So if you do have some
clear ideas of what you want to do, make sure they coincide with the dept's
strengths; this may mean modifying your statement for different
applications. 

I know that at some (mostly smaller) schools, the statement of purpose is
used to help make application decisions. But we get so many applicants
(600+) that undergraduate GPA and GRE scores (especially the generals) are
used for the first pass over the application pool, and recommendations
after that. It is much more important to get good recommendations ---
especially if you can get a recommendation from someone well-known, whose
will be familiar to people on the admissions committee --- than to have the
most interesting statement of purpose.

- ------------------------------------------------------------------------

Now that I've finished my phd, I can see what the statemtent of purpose
is for, but when I was applying for grad schools I also had no
idea what to say.  Professors want to work with students that
have interests similar to theirs so they can publish papers with them
and thereby get closer to getting tenure.  If you say that you're not
sure what you want to study in CS, they'll put your application on
a pile of "probably not"s and they'll look for applications where the
person says something more specifically related to their interests.
The thing you have to know is that most grad students change their
minds about what they want to specialize in once they get to grad school,
so it doesn't matter what you say you want to do.  A good trick would
be to look at a description of the professors and their interests
and make your interests compatible with theirs.  If you say you're
interested in parallel systems and no one at the school teaches about
parallel systems, you'll be rejected.

It would help if you could motivate your choice, like by saying
you want to work in parallel systems because you're fascinated by
making computers faster, and that also gives you a general interest
in systems stuff and optimization.

Of course, this is from a woman's perspective, and a male prof may
want you to brag a little, I don't know.  Why not ask a male prof of
yours what he'd look for?

Good luck--

=======================================================================

Hi,
	I was going to send you my purpose statement, but somehow I
have lost it :(  Anyway, my advice is to be as specific as you can
about what you want to do, (or just say what your favorite class was,
or that you found them all interesting), and *do* brag a bit.  Mention
any research you've done, or papers you've written, even if for a class
project.  Also, mention how you fit in at the university in question.  
Finally, give it to as many people as possible, even outside your field, 
to critique and give suggestions.  That helped a lot for me.

Good luck,

=========================================================================

(edited from a talk that was presented during the Graduate School 
Experience panel at the Grace Hopper Celebration of Women in Computing.
The purpose of the talk was to describe the admissions process and give
some ideas of what we look for in each element of the application.)

The personal statement is your one chance to tell the admissions committee
who you are and where you are going.  When judging statements, we look for
three things: preparation, fit with department, and writing ability.  With
regard to preparation, tell us what you are looking for in graduate school,
and how your prior experience fits in with your anticipated future.  With
regard to fit, express your current research interests (what led you to
graduate school?).  It is best for both the department and the student if
there is some match between the student's interests and the departments
research projects.  It is a good idea to do some research on each graduate
school's research projects and tailor your personal statement accordingly.
Statements that praise our department on its excellence in a topic where no
current research is going on raise a red flag to the committee, and these
applicants are generally rejected.  Finally, the statement is our only
indication of your writing and communication ability.  Proofread your
statement carefully!  (I can't emphasize this enough).  Poor grammar,
spelling, or other errors (such as referring to the wrong institution) can
cause rejection since applications are so competitive.  Another pitfall to
avoid is the overly generic personal statement, e.g.  "I am interested in
all areas of computer science and I want to study at your department because
of your excellent research reputation in computer science."

==========================================================================
NOTE: a syster asked if she could forward my query to her husband since
he advises students on how to write these statements.  I have included
his comments here because I found them very helpful.  I have also
included his e-mail, with his permission, as he says he'd be glad to
answer any of our questions.

 jeffrey.donnell@me.gatech.edu
- ------
Here's what Jeff said..

 "I work for the Mechanical Engineering School at Georgia Tech,
 where I teach technical writing.  The most important technical
 writing that our undergraduates do is--applications.  So I spend
 the fall term of each year helping students to develop statements
 of purpose--both for graduate school admission and for national
 fellowships like NSF, AAUW, etc.
 
 I work with engineers more than with computing folk, but faculty
 reviewers across the disciplines seem to want similar stuff.  So
 I think the stuff below should give some good starting points.
 
 First--What do we mean by Statement of Purpose?  The statements
 of purpose that I have seen tend to be brief documents--usually
 about 300 to 500 words.  They are part of the university's
 application packet, and they are often (but not always) the only
 essay you have to write for the admission application.  
 
 What if people want a Personal Biography instead of a Statement
 of Purpose?  There is no practical difference between these
 documents.  In fact, you will find that there are still MORE
 names for the document we are talking about: Biosketch is pretty
 standard terminology, as is Background summary...
 
 Second--Are there instructions for these things?  Application
 packets usually tell you to cover the following stuff:
 background, training, experience, honors, and the reasons for
 pursuing a specific degree program at [name of university].  
 
 What do these instructions mean?  
 What is your job? 
 What does this document really DO?

 Review committes use your personal statement mainly to match
 you with an advisor.  What you are doing is explaining 
 how you might fit into an advisor's projects.  So...
 review committees generally want you to answer questions in the
 Personal Statement:
 
 	1)   What do you want to study in [this school]?
 	     Why do you want to do *this* *here*? 
 	2)   Do you have pertinent experience?
 	3)   What is your motivation for doing this?
 	  
 Review committees need to see answers to these questions.  
 It is your job to provide answers.  
 
 Third--How should you think about it?  You should think about the
 stuff you say to interviewers (if you've had interviews); these
 are pretty much the same questions.  Or maybe you've gone to job
 fairs or grad school fairs where you pretty much keep repeating a
 90-second blurb describing your experiences and your goals.  
 
 ----------------------
 
 A few Details.  
 
 1) "What do you want to study, and why do you want to do it here?" 
 
 You should not try to answer this question alone.  You should
 start off by collecting research guides (or brochures or
 summaries) from the different departments where you will apply.
 You'll look through these things and you'll find summaries of
 ongoing research in the different areas that [that school]
 offers.  You'll find a few projects (and possible faculty
 advisors) that interest you, and you will ask yourself this
 question: "If I worked in this [area], and if I worked on chunks
 of these projects, what would I try to do on my own?"  
 
 The answer to this question should be a few sentences long, and
 it will form about a third of your Personal Statement.  
 
 2) "Do You have Pertinent Experience?"  Pull out your resume and
 describe--aloud--the projects that it hints at.  You don't need
 to talk about your hometown, and you don't need to talk about the
 other disciplines you sampled and abandoned.  You need to talk
 about the things you have done WHICH WILL HELP THEM TO MATCH YOU
 WITH AN ADVISOR.
 
 3) "What is Your Motivation for doing this?"  Why on earth would
 you want to put yourself through the trouble of grad school?  You
 need to answer this question.  Committee members know that people
 who really want the degree are people who finish, and these are
 the folks that committees want to recruit.
 
 ----------------------------
  
 Last--What to do with the essay.  
 
 When you've written answers to these questions, DO NOT MAIL THE
 ESSAY.  
 
 Instead, you should begin to show it to people.  It is called a
 personal statement, but you need to make this a very public
 document.  You need to show it to the people who will write your
 reference letters; you should also show them a resume.  You do
 this because it is your job to teach them what letter they should
 write.
 
 After you give them this information, you deserve to get some
 information from them, so you need to ask them some questions:  
 
 	a)  Is [x] a good school for your interests?
 	b)  Is [z] a good advisor--
 		hint--do z's students finish in less than 10 years?
 		      do z's papers have student coauthors?
 	c)  Is my plan (project) credible?
 	d)  Should I add/alter anything?
 	e)  Should one pursue additional issues on such a 
 		project?
 
 ------------
 
 You want to conduct this conversation with all of your
 references.  AFTER you get their comments and suggestions, you
 should edit your plan/or your document, you should distribute
 copies with the reference letter forms, and you can then mail off
 your package.
 
 Good luck everybody! 
 I hope this helps!"
==========================================================================

jmankoff@cs.oberlin.edu, csmc@cs.oberlin.edu