USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 34 of 80
upon their wives”).
The Supreme Court in Craig acknowledged that statistics and numbers have surface appeal. But the Court also knew it was “unrealistic to expect either members of the judiciary or state officials to be well versed in the rigors of experimental or statistical technique.” 429 U.S. at 204, 97 S. Ct. at 460. The easy manipulation of statistics by parties, and the possibility of being fooled by them, “illustrate[] that proving broad sociological propositions by statistics is a dubious business, and one that inevitably is in tension with the normative philosophy that underlies the Equal Protection Clause.” Id.
We view the dissent’s reliance on a figure of 99.96 percent as a prime example of the folly of relying on statistics in equal protection analysis. Dissenting Op. at 44–47, 58.[1] The relevant inquiry in this case is not what percentage of St. Johns students are transgender, but whether the challenged policy furthers the important goal of student privacy. And on that point, the dissent offers no statistics. Nor could it. When it comes to actual reported invasions of privacy
- ↑ The dissent defends its reliance on its 99.96 percent figure by pointing to instances where the Supreme Court recounted statistics outside of the context of an Equal Protection analysis. These cases are inapposite. See Dissenting Op. at 46–47 (citing Birchfield v. North Dakota, 579 U.S. __, 136 S. Ct. 2160, 2193 n.10 (2016) (Sotomayor, J., concurring in part and dissenting in part) (writing in a Fourth Amendment case); Cent. R.R. Co. of Pa. v. Pennsylvania, 370 U.S. 607, 610 n.2, 82 S. Ct. 1297, 1301 n.2 (1962) (recounting the number of railroad cars located on out-of-state railroads in the facts section of a case concerning capital stock tax)). No one is saying members of the judiciary are unable to do simple math. But the Supreme Court has time and again warned against using arithmetic to supply pat resolutions to sex discrimination cases.
34
