Utah v. Evans

GREENE, United States District Judge, concurring in part and dissenting in part.

I concur with the majority that the plaintiffs’ Census Act claim is justiciable. I disagree, however, with the majority’s conclusion that the use of imputation to arrive at apportionment numbers in the 2000 Census was not prohibited under 13 U.S.C. § 195, as interpreted by the Supreme Court in Dep’t of Commerce v. House of Representatives, 525 U.S. 316, 119 S.Ct. 765, 142 L.Ed.2d 797 (1999). For that reason I would rule that § 195 of the Census Act was violated and that the injunctive and declaratory relief sought by plaintiffs should be granted. My analysis is limited to a consideration of the statute, so I do not reach or discuss the APA or constitutional claims asserted by plaintiffs, and I would abstain from acting thereupon. The issue of disagreement presented by this dissent is whether the use of “imputation” in arriving at the final 2000 Census apportionment count constitutes “sampling” which is prohibited under § 195 of the Census Act. If so, the statute was violated. If not so, no violation occurred.

Judge Michael Murphy, in a very well written opinion for the majority in this case, has ruled that the use of imputation is not prohibited by the statute for the following reasons:

Sampling is the technique of determining the traits of the entire population by *1181collecting and analyzing data from a representative segment of that population. In hot deck imputation, on the other hand, there is no representative sample from which to infer the characteristics of the larger population. Instead, a single, fully enumerated housing unit is used to draw conclusions about a neighboring housing unit for which data is missing. That is, imputation is used to fill in missing data on the assumption that similar types of dwellings in the same geographic area will have similar characteristics. Because imputation is not “the statistical method known as ‘sampling,’ ” section 195 does not prohibit its use in the 2000 census.

Majority Opinion at 25 (emphasis added).1

The statute in question, as amended in 1976, provides:

Except for the determination of population for purposes of apportionment of Representatives in Congress among the several States, the Secretary shall, if he considers it feasible, authorize the use of the statistical method known as “sampling” in carrying out the provisions of this title.

13 U.S.C. § 195 (emphasis added).

The Supreme Court in House of Representatives ruled that “the Section maintains its prohibition on the use of statistical sampling in calculating population for purposes of apportionment” and that a plain reading of the “text of the statute” demonstrates that “the use of sampling for apportionment purposes” is prohibited. 525 U.S. at 339, 343, 119 S.Ct. 765. The court went on to conclude that “the Census Act prohibits the proposed uses of statistical sampling in calculating the population for purposes of apportionment.” Id. at 343, 119 S.Ct. 765.

Since the use of imputation was not directly before the court in House of Representatives, of course there was no ruling as to whether or not it is prohibited by § 195. However, as will be seen, the essence of the use of imputation in Census counting is that it is a form of sampling. As such, it ought to be struck down for the same reasons the proposed uses of statistical sampling were struck down in House of Representatives. No definition of sampling was set forth in House of Representatives, but for many years the Census Bureau has defined sampling broadly, and its basic meaning has been and is common knowledge. Hence, the “statistical method known as sampling” is no secret and it was aptly described by the Bureau in its 1997 report to Congress thusly:

In our common experience, “sampling” occurs whenever the information on a portion of a population is used to infer information on the population as a whole.... In laymen’s terms, a “sample” is taken whenever the whole is represented by less than the whole.

United States Department of Commerce Bureau of the Census, Report to Congress — The Plan for Census 2000, p. 23 (Revised and Reissued August 1997).

Prior to 1994, the Bureau studiously avoided the use of estimates to arrive at a *1182population count for purposes of apportionment. That prior history, including reference to testimony given by the Bureau in court proceedings and to Congress, was noted in the majority opinion written by Justice O’Connor in House of Representatives. See 525 U.S. at 340, 119 S.Ct. 765. The consistent past opposition by the Census Bureau concerning use of estimates in determining the apportionment count was explained by a Bureau representative in a court proceeding in 1981:

[S]ince 1790 the census enumeration has never been adjusted to reflect an estimated undercount and that in their opinion Congress by statute had prohibited such an adjustment in the figures used for purposes of Congressional apportionment.

Young v. Klutznick, 652 F.2d 617, 621 (1981) (emphasis added).

By the mid-1990s, as set forth in its 1997 Census Report to Congress, the Bureau’s position had changed and was in full support of the use of sampling, including the use of imputation. In that Report, under a general section entitled “Reliance on Sampling in Previous Censuses” and also under a subheading entitled “Use of Scientific Sampling to Increase Accuracy,” imputation and other forms of sampling were discussed. See United States Department of Commerce Bureau of the Census, Report to Congress — The Plan for Census 2000, p. 23.2 The Bureau’s reasoning for the use of imputation in Census apportionment counting was articulated by Kenneth Prewitt, then Director of the Bureau. He stated that “the apportionment number includes imputed Census records, that is people we have not talked to, but [which] we impute into the Census file.” Oversight of the 2000 Census: Hearing of the Census Subcommittee of the House of Government Reform Committee, at 2 (June 22, 2000). Mr. Prewitt further explained:

[I]n the apportionment number there will be a certain number of Census records which are put there through an imputed process. Those are not people with names. They are not people who filled out a form. They are people who our statistical processes lead us to believe by putting that Census record in there we have given the country a more accurate number. That’s an adjusted number.

Id. at 3 (emphasis added).

In his dissenting opinion in House of Representatives, Justice Breyer argued that statistical sampling should be allowed as a supplement in arriving at final population figures for purposes of apportionment. See 525 U.S. at 353, 119 S.Ct. 765. Justice Breyer then discussed imputation as an appropriate statistical procedure in use by the Census Bureau to “fill in gaps” in arriving at the final population count. Id. at 352, 119 S.Ct. 765.

Justice Breyer’s description of the way imputation is used to enhance the population count fits well within the definition of sampling used by the Bureau now and historically. Justice Breyer said:

When an enumerator believes a residence is occupied but is unable to obtain any information about how many people live there, the Census Bureau “imputes” that information based upon the demographics of nearby households. Imputation was responsible, for example, for adding 761,000 people to the Nation’s total population in 1980 and 53,590 people in 1990.

Id.

The majority in House of Representatives rejected Justice Breyer’s position *1183that sampling could properly be used as a supplement in population counting. The same reasoning should apply to the use of imputation as above described.

From the foregoing discussion, imputation is revealed as a form of sampling to establish and enhance a portion of the total population by imputing to nonresponding households and housing units the same number of occupants who live in certain occupied households.

Whether by the use of imputation apportionment numbers are enhanced by the use of a “statistical process” in order to arrive at “adjusted” numbers, as described by the Census Bureau; or whether the number is adjusted by a “statistical estimate” as described by Justice Breyer in order to “fill in gaps;” or whether “estimation techniques” are utilized to enhance population figures in the manner suggested by Justice Scalia in House of Representatives;3 or whether the number is enhanced so as to “fill in missing data on the assumption that similar types of dwellings in the same geographic area will have similar characteristics” as is described by the majority in this case; — the result is the same. All of these examples describe a form of sampling known as imputation in which statistical estimates are used to arrive at population apportionment. This crosses the line and violates § 195 of the Census Act. Accordingly, the use of imputation should be struck down as a prohibited form or subset of sampling under the reasoning of House of Representatives.

Manifestly, “sampling” and “imputation” in substance and effect are indistinguishable because both use a portion of the population to infer information concerning segments of the population in order to arrive at final figures concerning the population as a whole. •

For the reasons aforesaid, I respectfully dissent.

. In coming to its conclusion, the majority relies in part on dicta in Orr v. Baldrige, Slip. Op. (D.Ind. July 1, 1985), an unappealed district court case in which imputation, apparently for tactical reasons, was stipulated not to constitute sampling. Since that matter did not involve any dispute concerning imputation as related to sampling, the court's dicta was based on a non-issue that was neither briefed nor argued. In this regard, Justice Souter enunciated this admonition in Church of Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 113 S.Ct. 2217, 2247, 124 L.Ed.2d 472 (1993), as to such undisputed issues: "Sound judicial decisionmaking requires ‘both a vigorous prosecution and a vigorous defense' of the issues in dispute....”

. The Bureau’s "Plan for Census 2000" states: "The Census Bureau also has used a form of statistical estimation to adjust or correct its actual headcount....”

. Justice Scalia, joined by three other justices in House of Representatives, warned of the use of "estimation techniques" in arriving at population/apportionment counts:

To give Congress the power, under the guise of regulating the "Manner” by which the census is taken, to select among various' estimation techniques having credible (or even incredible) "expert” support is to give the party controlling Congress the power to distort representation in its favor. In other words, genuine enumeration may not be the most accurate way of determining population, but it may be the most accurate way of determining population with minimal possibility of partisan manipulation. The prospect of this Court's reviewing estimation techniques in the future, to determine which of them so obviously creates a distortion that it cannot be allowed, is not a happy one.

525 U.S. at 348-49, 119 S.Ct. 765. Justice Scalia's comments were not directed specifically to § 195 of the Census Act, but they are instructive because the very mischief he seeks to avoid in eschewing "estimation techniques” could constitute violation of the statute.