I'm not relying on handwaving and racial stereotypes here. There are plenty of measurements pointing out the difference in standardized tests. For example, this document[1] shows SAT scores and percentiles divided by race. Assuming I'm not misreading this, it shows that 23% of Asian students score a 1400+ on the SAT compared to 1% of black students with the same score. That's more than a 10x difference. That seems a little hard to believe, but wikipedia[2] also has some illustrative graphs that suggest 10x+ differences. For example, figure 1-9 shows the difference in perfect math scores by race.
I don't think it's really a stereotype but a well documented fact that Asians outscore white and black students both on standardized tests. I also think that tests are kind of like school. Tests are similar to how you are graded, they are about similar content, classes are meant to prepare you for tests, etc. It seems straightforward to me to say that if Asian students are good at the kind of standardized tests they encounter in school, we should also expect them to be good at the standardized tests that gate high achievement programs, and that we should expect them to enter and perform well in those programs.
I'm not familiar with any evidence showing the efficacy of test prep and I'd be really surprised if it was as large as the differences between groups that I've pointed to in the sources mentioned above. Perhaps you could share something to read on the topic?
Regardless, students who take test prep courses are probably a better fit for advanced classes that require students to do more work. Doing more work is how you prepare for tests and how you perform in the advanced classes. As I mentioned before, I think it is absolutely worthwhile to check and see if racism is happening here - and I think you could do that by checking what percentage of students by race pass the test, win the lottery, and are admitted. If there are unexplainable discrepancies between those steps, that would be good evidence of racism, the fact that different racial groups perform differently on the tests seems just like a recreation of a well known and oft-observed and documented phenomenon.
1 - https://collegereadiness.collegeboard.org/pdf/sat-percentile... 2 - https://en.wikipedia.org/wiki/SAT#Association_with_race_and_...
Yes, I agree that this would be evidence of racism. But you can't turn the argument around. The lack of any discrepancies does not mean that black or hispanic students aren't at a disadvantage, because the disadvantage might be something that lead to the worse test results in the first place.
Parental income or education could be a systemic disadvantage that is not racist. Or a lack of role models. Or different amount of support by families.
What you're describing seems less systemic and more naturally occurring. Nobody forces Asian families to put more emphasis on academics or black families to put less, that just tends to be the way things happen. If there is a system that encourages it I think it would only be the kind of "system" that could be anything.
Something like parental situation likely has a great deal to do with this and that's important to point out. For example, suppose that children raised by a single parent did worse academically, and suppose that 2/3 black families had only one parent compared to 1/5 Asian families. That's important to know because it means that to fix the problem we would need to encourage black people to stop creating single family homes rather than to force schools to stop educating white and Asian advanced students. I think this approach would result in a culture that spends less "D&I energy" on things like making sure school programs are exactly the right color and more on encouraging good behavior and explaining the problems of single parent families and how they can be avoided.
In any case, I don't really care much about the distinction between "natural occurring" or "caused by some system". Of course it's important to search for the underlying reasons for inequality, but you can still do things even if you don't know what exactly is ultimately causing the inequality.
If you discover that a school program only benefits some races, because your tests tend to select only certain races, you don't need to change everything about society to make sure all races have similar test results.
You could also just change the admission criterion -- and I'm not talking about quotas. Maybe personal interviews with teachers or social workers would work better for selecting students? Or maybe you should just select the best students from each class, to avoid putting students at a disadvantage if they happen to have a bad teacher? Or maybe you could even change the whole program in a way that doesn't limit it to a few lucky kids who managed to get selected by some arbitrary process?
For example, you could just offer advanced classes open to anyone after school! Just let the kids decide for themselves if they want to try some more challenging things, without first testing them to make sure they really are worthy of your time.
A 10x difference needs a bit more explaining than some handwaving about racial stereotypes.
If there's a program, and it only helps the high achievers of a group that is likely already very privileged, it's a good idea to ask if that program is really working as intended.
And no, I don't think that explaining results with some arbitrary test scores is enough. Tests results often strongly depend on how well students were prepared for them, so they might just exacerbate previous inequality.