Sotomayor, J., dissenting
is forced to reconstruct the record and conduct its own factual analysis. It thus relies on a single chart from SFFA’s brief that truncates relevant data in the record. Compare ibid. (citing Brief for Petitioner in No. 20–1199, p. 23) with 4 App. in No. 20–1199, p. 1770. That chart cannot displace the careful factfinding by the District Court, which the First Circuit upheld on appeal under clear error review. See Harvard II, 980 F. 3d, at 180–182, 188–189.
In any event, the chart is misleading and ignores “the broader context” of the underlying data that it purports to summarize. Id., at 188. As the First Circuit concluded, what the data actually show is that admissions have increased for all racial minorities, including Asian American students, whose admissions numbers have “increased roughly five-fold since 1980 and roughly two-fold since 1990.” Id., at 180, 188. The data also show that the racial shares of admitted applicants fluctuate more than the corresponding racial shares of total applicants, which is “the opposite of what one would expect if Harvard imposed a quota.” Id., at 188. Even looking at the Court’s truncated period for the classes of 2009 to 2018, “the same pattern holds.” Ibid. The fact that Harvard’s racial shares of admitted applicants “varies relatively little in absolute terms for [those classes] is unsurprising and reflects the fact that the racial makeup of Harvard’s applicant pool also varies very little over this period.” Id., at 188–189. Thus, properly understood, the data show that Harvard “does not utilize quotas and does not engage in racial balancing.” Id., at 189.[1]
- ↑ The majority does not dispute that it has handpicked data from a truncated period, ignoring the broader context of that data and what the data reflect. Instead, the majority insists that its selected data prove that Harvard’s “precise racial preferences” “operate like clockwork.” Ante, at 31–32, n. 7. The Court’s conclusion that such racial preferences must be responsible for an “unyielding demographic composition of [the]
