response 195 of 232:

Jan 28 16:20 UTC 2003

"or drop out" could be read as "can't afford tuition". 

response 199 of 232:

Jan 28 18:25 UTC 2003

re resp:189: If that were my position, I would say so.  Do you often 
make such assumptions about people?  You do realize you're just 
categorizing me like this:

   "He disagrees with me, and I *know* I'm right, so there must be 
   something wrong with, not just his arguments, but the man himself."

I wish you were comparing arguments instead of doing that.  This isn't 
supposed to be about defending *me*.  I think you usually do a better 
job defending your position than you've done recently in this item.  I 
think you need to examine why it is you don't respect me or what I 
say.  It's clear that you don't.

response 203 of 232:

Jan 28 19:54 UTC 2003

re resp:188, resp:190-192: I am assuming that equally prepared students
 of different ethnic backgrounds succeed on about the same level.  I  recognize
this may not be completely valid, but I have reasons for it.

I'm interested in collegiate sports, and am aware that athletic 
scholarship students don't succeed on the same level as other 
students.  That's true at U-M; it's true almost universally among 
scholarship athletes in revenue-producing programs (football, men's 
basketball and ice hockey).  Among Big Ten schools, the difference in 
graduation rates is in the ballpark of 20%.  

There are a lot of reasons why scholarship athletes don't do well in 
school.  The guys who transfer, leave school to go pro, fail out of 
school or just quit are counted as "not graduating".  Those who remain 
for 4 or 5 years have lengthy amounts of time for practice, they travel 
a lot for their sport, and they concentrate more on athletics than 
academics.

And, they don't have to meet the same academic requirements as other 
students.  They may very well have been passed through their classes in 
high school because they're star athletes.  It happens.

I don't want to start another argument about the value of an athletic 
education.  I want to establish that I'm aware of, and interested in, 
some students who haven't got the same academic background as the 
general student body.

NCAA statistics for 6 year graduation rates for all 1995 freshmen:
all student athletes: 60%
all students: 58%
black male student athletes: 43%
black male students: 34%
white male student athletes: 59%
white male students: 59%

   Source: http://www.ncaa.org/news/2002/20020930/active/3920n01.html

One might conclude that black male athletes come from a more similar 
background to white male athletes, than black students in general 
compared to white students, and so they have more similar performance.  
It's still not as good as white male students or white male student 
athletes.

I guess there are lots of explanations possible for these numbers.  
Keep in mind they count all athletes, not just revenue athletes, but 
here are some numbers from football and basketball:

Revenue athletes graduation rates from 1994:
white football: 62% (better than white students in general)
black football: 35% (better than black students in general)
white basketball: 51% (worse than white students in general)
black basketball: 28% (worse than black students in general)

I'm not sure if any of this supports my contention that academic 
success is probably more proportional to background and preparation 
than race, but it was fun looking it up.

response 205 of 232:

Jan 29 01:53 UTC 2003

I don't have a cite for this, but I remember hearing a few years ago about
a study showing that UM minority students tended not to do as well as white
students with identical high school records, SAT scores, and other qualifying
information.

response 207 of 232:

Jan 29 13:24 UTC 2003

re resp:206: Should "diversity" be just for admissions, or should 
granting of degrees be for people of different ethnic backgrounds as 
well?  I think any university should be giving primary consideration to 
who can graduate.  There's no point in admitting anyone who can't.

response 209 of 232:

Jan 29 17:22 UTC 2003

UM is undertaking a study of the factors related to graduation rates for
different demographics. They have lots of data, but are going over them
to determine which data are statistically significant. 

Re #207: the objective is to provide an environment in which all admitted
students have the potential to graduate without compromising standards. 

response 211 of 232:

Jan 29 18:04 UTC 2003

I doubt that, but there is an inescapable subjective element in the choice
of significance levels. 

response 213 of 232:

Jan 29 23:06 UTC 2003

Re #206:  If the actual result of AA is to give minorities about
half the proportion of degrees that the rest of the student body
gets, meanwhile convincing many non-minorities that minorities are
academically unprepared or even inferior, is that good?

response 215 of 232:

Jan 30 02:05 UTC 2003

Just about anyone who wants to earn a college degree can probably get 
one, but he has to select a college that matches his level of 
competence.  You are solving the problem as Solomon suggested (cut the 
baby in half).  Do you really believe that the student who drops out 
because he is inappropriately placed is a sacrifice worth making?

response 217 of 232:

Jan 30 04:47 UTC 2003

Re #214:  Let's play a little bit with numbers, just for curiosity's sake.

Suppose that the current AA system graduates 34% of AA-qualified
minorities and 59% of others.  This is a dropout rate of 66% for
the AA group and 41% for others.

Change this so that AA-qualified minorities no longer get admitted
to institutions for which they would not otherwise qualify.  They
go to less-rigorous institutions instead, and are replaced with
non-AA students who would have made the cutoff otherwise.  Suppose
that this broader group would have a 57% graduation rate.

If the AA-qualified are otherwise no different from the rest of
the population, their graduation rate would rise from 34% to 57%
(23% increase).  That is a 2/3 improvement.  Meanwhile, the broader
population's graduation rate falls from 59% to 57% (a 3.4% drop).

If AA students make up 10% of the population, the total graduation
rate with AA is .34 * .1 + .59 * .9 = 56.5%, while afterwards the
total graduation rate is 57%.  This is slightly better as a whole,
while being far better for the AA group. 

It appears that the race-sensitive admissions program could easily
be *worse* for minorities (in terms of degrees awarded), and worse
for the non-minorities - the exact opposite of the intended result!
Of course, this analysis is highly sensitive to the guesstimate of
the graduation rate of the broader population.  If the aggregate
graduation rate of a student body selected without AA is 55%, the
effect of AA is to *increase* majority graduation numbers while
hurting minorities.  This is exactly the sort of discrimination AA
was intended to solve.... wasn't it?

response 219 of 232:

Jan 30 07:01 UTC 2003

Realize also that no one is forcing admitted students to attend UM
(most don't), and it is sure to be common knowledge among all demographics,
what are the "chances" at any school they consider. Only an *opportunity* is
being offered. So all that numerology about % graduating is largely
beside the point in regard to AA. It *is* of importance in making the
best use of resources avalable to the University, but that is a different
issue.