Federal appeals court once again tosses lawsuit over exam-school admissions policy aimed at increasing Black and Latino enrollment

A federal appeals court yesterday rejected, again, a bid to block Boston Public Schools from changing its previous grades-and-test-score criteria for admissions to Boston Latin School, Boston Latin Academy and the O'Bryant School.

A group of mostly White parents calling themselves the Boston Parent Coalition for Academic Excellence had argued the new plan as racially biased because it reduced the percentage of White and Asian-American students at the three schools.

But in its second ruling on the matter, the US Court of Appeals for the First Circuit in Boston basically said the parents had it backward: That because the new system, which included Zip codes as a criterion, worked as planned, it actually reduced racial disparity at the schools by increasing the number of Black and Latino students, who had long made up a smaller percentage of enrollment than their overall numbers in BPS and the city as a whole.

[T]he Coalition offers no evidence that geography, family income, and GPA were in any way unreasonable or invalid as selection criteria for public-school admissions programs.

And to allow the old racial/ethnic percentages to prevail would only invite yet another lawsuit, this time from Black and Latino parents, because they could show that the system was gamed against them, the court said.

It continued that even in its recent Harvard admissions affirmative-action decision, six of the justices - including three in the majority - said that schools could craft admissions policies aimed at ending racial disparities that did not involve specific questions related to an individual student's race.

Justice Gorsuch, and indeed plaintiff Students for Fair Admissions itself, identified use of socio-economic status indicators -- i.e., family income -- as a tool for universities who "sought" to increase racial diversity.

The court also dismissed what it called some statistical "sleight of hand" that the Boston group's attorneys used in their arguments:

The Coalition contends that the Plan, even when measured against a process of random selection, had a disparate impact on White and Asian applicants. To make this argument, the Coalition first notes that the overall acceptance rate for applicants for the 2021–2022 school year was 58.5%. And it posits that a random distribution would result in an even application of that 58.5% rate across each zip code. The Coalition then isolates certain zip codes where the population was either "predominantly" (as in 55% or greater) White/Asian or Black/Latinx, and juxtaposes those zip codes' respective acceptance rates under the Plan with those under a hypothetical 58.5% comparator. Following this logic, the Coalition concludes that the Plan resulted in 66 fewer than expected spots allocated across ten predominantly White/Asian zip codes, and 57 more spots across seven predominantly Black/Latinx zip codes. Using this same data, the Coalition also argues that because the average GPA of the admitted students from the predominantly White/Asian zip codes was higher than that from the predominantly Black/Latinx zip codes, the Plan made it disproportionally more difficult for White and Asian students to gain acceptance.

In our view, this backfilled analysis -- crafted by counsel in an appellate brief -- falls woefully short of the mark. The analysis uses GPA data from only ten of the twenty zip codes that the Coalition identifies as "predominantly" White and Asian. It also neglects another two zip codes where, ostensibly, there was neither a predominantly White/Asian nor Black/Latinx population under the Coalition's definition. And all the while, the Coalition never explains why 55% should be the relevant threshold, nor why aggregating populations of separate racial groups is methodologically coherent.

Moreover, the Coalition's analysis rests on a sleight of hand. It counterfactually assumes that if White/Asian students comprised 55% or more of the students in a given zip code, then every marginal student in that zip code who just missed out on acceptance was also White or Asian. Suffice it to say, there is zero evidence for this assumption. The bottom line remains the same: White and Asian students respectively made up approximately 16% and 7% of the eligible school-age population and 31% and 40% of the successful applicants. Use of the Plan caused no relevant disparate impact on those groups.

Yes, but what about the fact that three of the seven members of the School Committee let their inner racists out - the committee chairman mocking Chinese names, the two Hispanic members snorting in texts about "Westie Whites?"

The court allowed as how all that was quite reprehensible, but:

We recognize that the text messages evince animus toward those White parents who opposed the Plan. But the district court supportably found as fact that the added element of animus played no causal role that was not fully and sufficiently played by the motive of reducing the under-representation of Black and
Latinx students.