I would bet differently, having checked the evidence.
http://www.businessweek.com/magazine/content/03_27/b3840045_...
http://www.educause.edu/ir/library/pdf/ffp0621.pdf
http://www.educause.edu/ir/library/pdf/ff0615S.pdf
http://www.tcf.org/Publications/Education/carnrose.pdf
http://www.tcf.org/Publications/Education/kahlenberg-affacti...
http://harvardmagazine.com/2005/05/a-thumb-on-the-scale.html
http://www.theatlantic.com/doc/200511/financial-aid-leveragi...
http://www.thecrimson.com/article.aspx?ref=510012
http://www.equaleducation.org/commentary.asp?opedid=1240
http://www.jkcf.org/assets/files/0000/0084/Achievement_Trap....
http://www.reason.com/news/show/123910.html
The most disadvantageous position to be in, in United States college admission, is to have little money. Low-ability, high-income college applicants fare much better in the application process (they end up with more and better choices of colleges) than high-ability, low-income students.
P.S. The submitted article, contrary to the HN submission title, refers only to grade averages at college, but I think there is also considerable evidence, for which I have fewer convenient citations,
http://www.amazon.com/Wad-Ja-Get-Grading-Game-American-Educa...
that higher-income, lower-ability high school students tend to enjoy better grades than their lower-income, higher-ability classmates. Teachers respond to signs of wealth.
Agreed. I am not a fan of the testing strategy where the average is a whopping 30% and a curve is used to bring scores back within the normal letter grade buckets. Testing relative performance at this level encourages students to study exclusively "to the test". It reaches the point where gaining inside knowledge about the possible contents of the next exam by hounding professors/TA's or analyzing a friend's exams from last semester takes precedence over comprehensive learning, since the difference between a C and an A can be as tiny as predicting the presence of one particular question and nailing it from rote memorization. Such a deep curve also discourages students from studying collaboratively, since getting a good grade relies on X other students getting a worse score.
Having been to a private liberal arts college, the environment is extremely rigorous. You can get the same sort of thing at state schools, but you have to be in the honors programs. I'd be interested to see if there are honors-track programs at the larger public schools that show the same "grade inflation." But because the public school is self-contained, the effect, that students in some classes consistently receive higher GPAs, is hidden.
* I hate the term because it is so often used as if to imply that if the students at a university do too well, the explanation cannot be the merit of the students, but that the courses are necessarily too easy. If getting 95% on an exam represents a high level of learning/knowledge/work, and all of the students score 95%, they should all get A's. Professors sometimes attempt to combat "grade inflation" by giving ridiculously hard exams (coughAndrew Ngcough), but IMO all this does is bring to the fore uninteresting factors like how well one performs under time pressure, or how quickly one can perform arithmetic, or whether one has a photographic memory.