[CTcard]

Here's the actual table from http://www.cpec.ca.gov/StudentData/DiversityReport.ASP:

[table][tr][td] Ethnicity	Degree Recipients (Undergraduate)	State Population 18 to 24
Asian/Pac Is	34.64%	11.55%
Black	3.04%	7.79%
Filipino**	4.83%
Latino	14.53%	43.51%
Nat Amer	0.60%	1.05%
Other	2.37%	2.57%
White	39.99%	33.54%
		** Population numbers for Asian/Pacific Islander include Filipino.
		Data Generated on Wednesday, June 03, 2015 at 5:01:36 PM

[/td][/tr][/table]

Have missed most of this thread while off in China with very limited internet access. The issue affects my family and my professional life, so have thought about it a lot.

I think that Stanford considers itself more of a national university than a California university. So Stanford probably considers national numbers more relevant.

Trying to get consistent numbers for racial demographics is difficult, with all sorts of ambiguities in definitions, what you are going to count, how people respond to surveys, etc. Here's an attempt to flesh out some numbers for the USA and Stanford - using a wikipedia article taken from the 2010 census and Stanford's reported common data set (required by Dept of Education). The comparisons are imperfect.

15-19 year olds:
[White, non-Hispanic  44.23%]*  [estimates - see below]
[Hispanic                    23.2%]*
White                          66.35%
Black                          15.64%
Asian                            4.34%
Native American            1.2%
Pacific Islander              0.21%
Mixed                          12.27%


Stanford must report related numbers in the "common data set". In 2014-15 the numbers were:
White (non-Hispanic) 37.5%
Hispanic                    16.1%
[White, approx          52.8% ]*     
Black                          5.9%
Asian                          19.8%
Native American          1.1%
Pacific Islander            0.4%
Mixed                        10.7%
Unknown                      0.3%


So, IF Stanford's goal is to have a student distribution that matches the nation by these broad categories, then only Asians are over-represented and every other group is under-represented - Blacks the most.

Any system you devise to try to make your university demographics more like the country than other admissions measures would give you, by definition gives preference to some and anti-preference (i.e. discrimination) to others. Those hurt the most are going to be those with the otherwise best measures. At the moment that's Asian-Americans. 

Now Stanford and it's brethren quite explicitly refuse to define just what measures are used for admissions in a quantitative manner - partly because they don't have such a quantitative approach and partly to avoid facing this issue head on - and almost certainly with an eye towards law suits as well.

So yeah, Stanford discriminates against Asian-Americans if you believe GPA and test scores should dominate admissions decisions. Stanford probably discriminates against Asian-Americans if you think everything but racial identity should be considered in admission, though this would be impossible to explicitly determine. But one way or another, discriminating against somebody is the cost of affirmative action. That's the argument against affirmative action - and at the least should be a warning to use it carefully.

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


*[Given that Hispanics are not a racial group, and thus don't factor into these census categories, these first two categories are approximated to match the common data set for universities.  Apparently 95% of hispanics are white by the designation above. Hispanics are something like 1/3 of the whites in the country as a whole, but I cannot find a number that exactly corresponds to the 15-19 year old category above. If both the 1/3 and 95% were to hold, then that would make Hispanics about 23.2%. ]

[CTcard]

Here's the actual table from http://www.cpec.ca.gov/StudentData/DiversityReport.ASP:

[table][tr][td] Ethnicity	Degree Recipients (Undergraduate)	State Population 18 to 24
Asian/Pac Is	34.64%	11.55%
Black	3.04%	7.79%
Filipino**	4.83%
Latino	14.53%	43.51%
Nat Amer	0.60%	1.05%
Other	2.37%	2.57%
White	39.99%	33.54%
		** Population numbers for Asian/Pacific Islander include Filipino.
		Data Generated on Wednesday, June 03, 2015 at 5:01:36 PM

[/td][/tr][/table]

Have missed most of this thread while off in China with very limited internet access. The issue affects my family and my professional life, so have thought about it a lot.

I think that Stanford considers itself more of a national university than a California university. So Stanford probably considers national numbers more relevant.

Trying to get consistent numbers for racial demographics is difficult, with all sorts of ambiguities in definitions, what you are going to count, how people respond to surveys, etc. Here's an attempt to flesh out some numbers for the USA and Stanford - using a wikipedia article taken from the 2010 census and Stanford's reported common data set (required by Dept of Education). The comparisons are imperfect.

15-19 year olds:
[White, non-Hispanic  44.23%]*  [estimates - see below]
[Hispanic                    23.2%]*
White                          66.35%
Black                          15.64%
Asian                            4.34%
Native American            1.2%
Pacific Islander              0.21%
Mixed                          12.27%


Stanford must report related numbers in the "common data set". In 2014-15 the numbers were:
White (non-Hispanic) 37.5%
Hispanic                    16.1%
[White, approx          52.8% ]*     
Black                          5.9%
Asian                          19.8%
Native American          1.1%
Pacific Islander            0.4%
Mixed                        10.7%
Unknown                      0.3%


So, IF Stanford's goal is to have a student distribution that matches the nation by these broad categories, then only Asians are over-represented and every other group is under-represented - Blacks the most.

Any system you devise to try to make your university demographics more like the country than other admissions measures would give you, by definition gives preference to some and anti-preference (i.e. discrimination) to others. Those hurt the most are going to be those with the otherwise best measures. At the moment that's Asian-Americans. 

Now Stanford and it's brethren quite explicitly refuse to define just what measures are used for admissions in a quantitative manner - partly because they don't have such a quantitative approach and partly to avoid facing this issue head on - and almost certainly with an eye towards law suits as well.

So yeah, Stanford discriminates against Asian-Americans if you believe GPA and test scores should dominate admissions decisions. Stanford probably discriminates against Asian-Americans if you think everything but racial identity should be considered in admission, though this would be impossible to explicitly determine. But one way or another, discriminating against somebody is the cost of affirmative action. That's the argument against affirmative action - and at the least should be a warning to use it carefully.

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


*[Given that Hispanics are not a racial group, and thus don't factor into these census categories, these first two categories are approximated to match the common data set for universities.  Apparently 95% of hispanics are white by the designation above. Hispanics are something like 1/3 of the whites in the country as a whole, but I cannot find a number that exactly corresponds to the 15-19 year old category above. If both the 1/3 and 95% were to hold, then that would make Hispanics about 23.2%. ]