Utah v. Evans, 536 U.S. 452, 56 (2002)

Page:   Index   Previous  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  Next

Cite as: 536 U. S. 452 (2002)

Opinion of Thomas, J.

tives from Indiana to Florida, id., at 46, 224, making the year 2000 census at least the second time that its use has changed apportionment.12

At the earliest, status imputation was used in the year 1990 census, although there is some dispute as to whether it was even used then. Id., at 45-46, n. 4; but see id., at 223 (stating that "the 1990 imputation procedures continued the prior practice of using household size imputation and occupancy imputation but added status imputation"). Regardless, it apparently had no impact on apportionment. See id., at 45-46, n. 4. In the year 1990 census, the Secretary specifically decided against using a different form of estimation. The "Secretary's administrative decision declining to make an adjustment observed that '[t]he imputation scheme used . . . [was] based on a series of assumptions that are mostly guesswork.' " Brief for Federal Petitioners in Wisconsin v. City of New York, O. T. 1995, Nos. 94-1614 etc., p. 8. The Secretary even noted that "large-scale statistical adjustment of the census through [this method] would 'abandon a two hundred year tradition of how we actually count people,' " and that "statistical adjustment of the 1990 census might open the door to political tampering in the future." Wisconsin v. City of New York, 517 U. S. 1, 10-12 (1996).

Though different in kind, our recent history of experimentation with census-taking methods bears similarity to the various preratification estimates and enumerations. While I would not speculate about the Bureau's decisionmaking process, it is quite evident that the Framers, aware that the use of any estimation left the door open to political abuse, adopted the words "actual Enumeration" to preclude the availability of methods that permit political manipulation.

Additionally, hot-deck imputation is properly understood as an estimation, which by definition cannot be an actual

12 The Bureau states it "no longer has data available to determine whether count imputation affected apportionment in the 1960 or the 1970 Censuses." App. 224.

507

Page:   Index   Previous  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  Next

Last modified: October 4, 2007