COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "let.pub" source file
C00007 00003	∂AIL$John McCarthy↓Professor and↓Chairman of 1976 Admissions$$Autumn 1975∞
C00022 ENDMK
C⊗;
.require "let.pub" source file;
.<<
.FONT 1 "basl30"; TURN ON "%";
.FONT 2 "BASI30";
.FONT 3 "basb30";
.FONT 4 "sta200";
.FONT 5 "ngb25";
.>>
.FONT 6 "SUP";
∂AIL$John McCarthy↓Professor and↓Chairman of 1976 Admissions$$Autumn 1975∞
Dear Propective Transfer Student:

Several students in the past have developed an interest in Computer Science
after being admitted to another graduate department at Stanford, and in some cases
they have transferred to Computer Science and done very well here.  Our policy is to
consider such transfer possibilities at the same time we evaluate admission
applications from outside Stanford; all entry to Computer Science comes essentially
through the same door.  So if you want to transfer, the procedure you should follow is
very much like what you already went through to get admitted to your present
department.  Here's what to do:

1. Obtain a petition form from the information window of the Registrar's Office
in Old Union.

2.  In addition to the petition, several letters of recommendation (up to three)
should be sent to this Department.  It is most desirable if the recommendations
can be obtained from Computer Science faculty members.  We expect that you will
have taken several graduate-level courses in our department, so that we have
evidence of your interest and ability and so that you can make an informed
decision about whether you really want to join up with us.

3.  Prepare a "statement of purpose," two hundred words or so, telling about
your chief interests and motivation.

4.  Fill out the special yellow C.S. application form enclosed.  It is especially
important that you specify here which program you are applying for:  Our
department primarily offers a program leading to the Ph.D., and also a separate
program leading to "Master of Science in Computer Science: Computer Engineering
(CS:CE)."

5.  %3Most important:%1  Your application (i.e., the petition, recommendations,
statement of purpose and yellow form) must be completely in our hands by
February 1, 1976.  We shall then attempt to do our part by letting you know our
decision before April 1, 1976.  These dates are geared to national policies
for awarding fellowships.  (There is an exception to this rule, namely that
we will continue to accept applications for admission to the terminal CS:CE
program until June 1, because no financial aid is offered to students in this
program.)

We will be able to admit only a total of 20 students to our Ph.D. program next
spring, because of the size of our department, and last year we had over
260 applicants.  So we will be able to say yes to only about 1 in 13 of those
who apply, and if you are not accepted it doesn't mean that we think you aren't
qualified.  If you still wish to transfer after reading all this, 
take heart:  Several of
our present students are living existence proofs that it can be done and with
excellent results.

.sgn
∂AIL$John McCarthy↓Professor and↓Chairman of 1976 Admissions$$Autumn 1975∞
Dear Prospective Applicant:

We are pleased that you have shown interest in admission to the Computer Science
Department.  The necessary application forms are enclosed with this letter.  In
order to help you prepare the most advantageous application, I'd like to describe
what we will do after we receive it, assuming that you want to enter our Ph.D.
program.

Everything sent to us relating to your application will be assembled into one
folder, and this folder will be studied by three members of our Admissions 
Committee.  The committee is composed of both faculty and student members, and
an attempt will be made to match your folder with readers who share your
major interests.  Each reader will carefully evaluate (1) the personal
recommendations, (2) your statement of purpose, (3) any examples of your work
that are enclosed, (4) the relevant courses you have taken and grades achieved,
and (5) the Graduate Record Examination scores.  We weigh the data in approximately
this order (i.e., recommendations count most).

Each reader will compare your folder with those of people who have been accepted or
nearly accepted in prior years, and if your application is sufficiently promising
it will be assigned a `score'.  We will take these scores and normalize them
in various ways to account for different readers' scaling techniques, and we will
sum the normalized scores.  The resulting magic numbers will be sorted in several
ways, thus giving some order to our subsequent consideration.  We won't blindly
use the magic numbers as an acceptance criterion, of course: At this point the
best folders will be looked at again by the committee as a whole.  And we will
finally come to a decision by late March.

In other words, we don't believe in automation so much that we would trust a
computer to evaluate the applications; but we do use the machine to help with
the things that can be done mechanically.

Last year we  received about 260 applications for admission to the Ph.D.
program.  Our department has about 15 faculty members with the responsibility
of teaching courses and guiding graduate students to advanced degrees.  For
a faculty this size, we should probably have about 60 grad students, but in fact
we have over 90.  All this adds up to a severe limitation on the number of
people we can admit into the "pipeline" each September, and we are forced to
turn down many extremely well-qualified students.  At present we intend to have
15 or so new students each year, so we will be able to accept only 20 applications.

In other words, we will be able to say yes to only about 1 in 13 of those who apply,
and if you are not accepted it %3doesn't%1 mean that we think you aren't qualified.

On the brighter side, if you are accepted there is a fair chance that you will not
have serious financial worries.  At least, in recent years almost all of our
full-time Ph.D. students have either won fellowships or we have been able to
scrape up some money to help them out, and we hope this good fortune will
continue.  (Please do not interpret this paragraph as an encouragement to avoid
applying for fellowships, of course, since our funds are limited and we can't
guarantee anything.  Many indicators point to the possibility of hard times ahead.
We love to hear that our admittees have won NSF fellowships.)  

Here are some more specific pointers to help you fill out your application.

1.  Make your statement of purpose as brief as you can.  Two hundred words or so
should be enough.  We'd like to know your chief interests as well as your
motivation; emphasize your plans, not your ideals.

2.  Take the Graduate Record Examination in October if possible, and by December
13 at the very latest.  You must take the "aptitude" test (verbal and
quantitative) and also an "advanced" test for some area other than engineering 
in which you feel competent.  For example, you might take the Advanced Math test,
or Advanced Physics, etc.  If for some reason you haven't been able to take the G.R.E.
by December 13 you will have to submit your application without it (since
the January scores will reach us too late); in that case a statement
explaining why you could not take the G.R.E. might keep your application alive.
We know that it is somewhat of a hardship for applicants to take these tests, but
we do want to have everyone's scores because it is usually the only source
of information that spans university boundaries.  Of course if you have already
taken the G.R.E. in prior years you may simply have the scores forwarded to us.

Perhaps you wonder what kind of G.R.E. scores are typical for the students in
our department.  Last year the verbal scores (percentiles) for admitted
applicants were

	20 32 67 77 77 85 87 89 92 95 96 96 98 98 98 98 98 99

(median raw score 680) and the quantitative scores were

	80 95 96 97 98 98 98 99 99 99 99 99 99 99 99 99 99 99 99

(median raw score 810).  So you can see that they aren't all 99s, 
and that we aren't terribly verbal; but
low G.R.E. scores make admission somewhat less likely because of the limited
number of possible acceptances.  The very low verbal G.R.E. scores were mostly
from students whose native language was not English.

3.  Choose people who know you, when you select your personal references, preferably
those most familiar with your recent academic or industrial work.

4.  Do not feel hesitant to apply because of little previous training in computer
science.  Unfortunately there still aren't very many undergraduate computer
science curricula of outstanding quality, and we almost prefer having our
students come from other disciplines; about half of our entering students
generally have a strong background in mathematics or physics, with only enough
experience of computer science to be aware of its exciting challenges.

5.  Be sure to fill out the special yellow application form for the Computer
Science Department, since we use it to simplify our data processing.  On this form
you must indicate which program you are applying for:  Our department primarily offers
a program leading to the Ph.D., and also a separate program leading to "Master of 
Science in Computer Science: Computer Engineering (CS:CE)".  A few candidates
are sometimes also admitted to a terminal M.S. in C.S. progam, as described
in Stanford's catalog %2Courses and Degrees 1974-75%1. However, our department is 
primarily oriented towards the Ph.D. degree, and you should circle one of the
Master's programs on the yellow form %3only%1 if (a) you have your own financial support
(e.g. by the Honors Co-op Program sponsored by local industry, or by a scholarship
from a foreign government); and (b) you do not intend to go on for the Ph.D.
degree.  (If you do want to work toward a Ph.D. but you apply only for M.S., you
must apply again for the Ph.D. later; and I'm afraid that none of the
people who made such transfer applications last year were accepted, in spite
of their high qualifications, because of the limited number of Ph.D. places.)
Acceptance to the M.S. in C.S. program is generally made only under rather 
special circumstances, e.g. for a foreign student with governmental support
to study abroad for a year or two.  If you want a Master's program that prepares
you for industrial or teaching positions, apply for M.S. in CS:CE.

6.  The University may send you information about financial aid, regarding the
"GAPSFAS" form requiring detailed statements of parents' finances, etc.
%3Don't%1 bother to read this, because it doesn't really apply to our department.
All that we look at is the yellow form where you say you do or do not desire
financial support.  (And your financial status does not influence our decision
with respect to Ph.D. admissions.)

7.  Please limit your personal communication with us to very urgent questions or
very special circumstances.  If you are accepted you will receive a lot of
additional information about housing, etc., but please don't ask for this now
or we'll not have as much time to study the application folders.

8.  %3Most important:%1  All applications for admission %3must%1 be in our hands
by January 15, 1976.  This means you should distribute the personal recommendation
forms before the end of 1975.  We shall then attempt to do our part by letting
you know our decision before April 1, 1976.  These dates are geared to national
policies for awarding fellowships.

There is an exception to this rule, namely that we will continue to accept applications
for admission to the Master's programs until June 1, because no financial aid is
offered to students in this program.  Admission to Master's programs is usually 
not decided until mid-June.

We look forward to receiving your application at an early date.

.sgn