Average ACT/SAT and Class Ranks/GPA for African Americans at Ivy League Schools

<p>Does anyone have an estimated range for this?</p>

<p>I highly doubt that stats like this are kept – it’s called plausible deniability.</p>

<p>You can find stats separated by race in various studies. For example, a couple days ago I referenced a Duke study in one of my posts, which listed the following average stats by race for Duke undergrads (data is from several years ago):</p>

<p>SAT (M+V):
Asian – 1457
White – 1416
Hispanic – 1347
Black – 1275</p>

<p>**HS Achievement<a href=“ranked%20by%20admissions%20on%20scale%20of%201%20to%205”>/B</a>:
Asian – 4.6
White – 4.2
Hispanic – 4.0
Black – 3.7</p>

<p>**HS Curriculum<a href=“ranked%20by%20admissions%20on%20scale%20of%201%20to%205”>/B</a>:
Asian – 4.9
White – 4.7
Hispanic – 4.7
Black – 4.3</p>

<p>**App Essays<a href=“ranked%20by%20admissions%20on%20scale%20of%201%20to%205”>/B</a>:
Asian – 3.5
White – 3.4
Hispanic – 3.2
Black – 3.1</p>

<p>**App LORs<a href=“ranked%20by%20admissions%20on%20scale%20of%201%20to%205”>/B</a>:
Asian – 3.9
White – 3.8
Hispanic – 3.5
Black – 3.5</p>

<p>**Low Income<a href="$50k/yr">/B</a>:
Black – 32%
Hispanic – 22%
Asian – 16%
White – 8%</p>

<p>^Links to source(s) please.</p>

<p>Hmm
I read online ivy ranges averages for AA; CCDPB: 1800-2000
HYP: 1900/2000-2100</p>

<p>Again, sources please. Sorry, but after 30 yrs doing research, just giving a number doesn’t hold any credibility without the original data source.</p>

<p>

<a href=“http://public.econ.duke.edu/~psarcidi/grades_4.0.pdf[/url]”>http://public.econ.duke.edu/~psarcidi/grades_4.0.pdf&lt;/a&gt;&lt;/p&gt;

<p>Bump</p>

<p>There is a problem with a study like this – you have to account for compounding factors such as athletics. Example – if the average SAT for athletes is in the 1200s – what percentage of the recruited athletes is African American versus white – so the lower SAT score may have more to do with the status of a recruited athlete than the race of an applicant. I am not saying this is true for Duke’s study – just that the variables need to be controlled for.</p>