Not that I read, however, that still doesn't explain the discrepancy between non-asian minorities and the other groups

"Non-Asian minority" is really really weird to me. We're just... making that distinction for some reason. Like what they don't count?

▲

aaplok 3 days ago | parent [-]

From the article:

> Research in the 1990s found that in California, Latino students scoring above the 90th percentile on standardized tests had only about a 50% chance of being placed in college preparatory classes, while Asian and white students with similar scores had more than a 90% chance.

The data is split by community ( Black, Latino, White, ...). Whites and Asians fare the best (see also their fig. 6.1). Kids from other communities are prevented from getting the same opportunities as kids from these two communities performing equivalently.

So it is not that Asian kids don't matter, it is that the data indicates that Asian kids already seem to be treated reasonably well.

▲

briangriffinfan 3 days ago | parent [-]

It just seems arbitrary to me. Split the groups of people into "good results" and "bad results" and then treating everyone in both groups the same seems so reductive as to not just be useless but convince me there's some kind of background reason for doing so.

▲

aaplok 3 days ago | parent [-]

They did not treat "results" as a binary "good" vs "bad" variable. In the article they take pains to justify their methodology in assessing students' chances of success in 8th grade algebra as a continuous variable:

> EVAAS is a statistical tool that could predict with remarkable accuracy which students would succeed in advanced courses.

Then

> The table demonstrates a strong correlation between EVAAS predictions and actual student performance. Students were grouped into probability ranges based on their predicted likelihood of success in 8th grade Algebra I, and their actual performance was then tracked.

So both the input (EVAAS predictor) and output (success in 8th grade algebra) are continuous variables (as shown in the first and last columns of the table mentioned in the quote), and they use this strong correlation to study access to 8th grade algebra against non-academic factors by using the EVAAS predictor as a control variable. I am not a professional statistician but honestly it looks pretty solid to me, or at least far more rigorous than most education science I have come across.

What they are then saying is that if you control for probability of success, Black and Hispanic children are significantly less likely to be admitted in 8th grade Algebra than White and Asian students. Looking at racial differences is not a particularly contentious way of studying bias in American society.

Of course they could have looked at other factors: economic and social class, gender, etc. This is the tragedy of sociology: you can't study social bias without being accused of introducing bias in how you study it...

▲

briangriffinfan 2 days ago | parent [-]

From my perspective this is like doing a study on how much the average person can bench and then concluding that society is unfair towards women because they bench less. That is to say I just don't take it for granted that the conclusion is "society is unfair (in this particular regard)."

▲

aaplok 2 days ago | parent [-]

You mean that you believe that Black and Hispanic kids are being blocked from taking Algebra in middle school because Blacks and Latinos are bad at Algebra compared to Whites, just as women cannot bench press as much as men?

Surely you can't be surprised that not everyone shares that belief??

	▲	briangriffinfan 2 days ago \| parent [-]
		I don't know the reason for the disparity, but I do know we've spent... how many decades at this point? trying and failing to correct for it. At some point I feel compelled to consider the possibility that our scope is wrong.