Early next week the new commerce secretary was to face the politically charged decision whether or not to adjust, by so-called statistical sampling, the 2000 census report to include millions of people left uncounted by the recent population tally. The issue of statistical sampling continues to spark heated political debate; advocates of the practice, mainly Democrats, say adjustments are the only way to redress what they say is the disenfranchisement of an estimated 3.4 million Americans, primarily minorities and poor people living in urban areas, who went uncounted in the last census. The plan's Republican opponents counter that complex statistical formulas increase the possibility of faulty numbers.
Given the political firestorm that would undoubtedly follow such a decision, Evans was reportedly approaching this announcement with some trepidation. Thursday, the Census Bureau made Evans' life a whole lot easier, advising the secretary in no uncertain terms to reject any revised numbers. There is no evidence, the bureau's director told Evans, that adjusted figures would be any more accurate than the original numbers. Evans' final verdict, his aides insist, is not set in stone and until it is, civil rights leaders and Democrats will pepper his office with statements opposing the bureau's recommendation.
But the fact that the Census Bureau is advising against statistical sampling certainly gives Evans a degree of political cover. And beyond its symbolic meaning, Evans' decision, which is still officially on hold until next week, will have practical political repercussions as well ones that stand to benefit the Republican party. Any ruling which applies only to districting, not the apportionment of federal money will mean that voter rolls in congressional districts remain tied to unadjusted census figures. If the adjusted numbers were used, Democrats and Republicans agree, Democrats could conceivably gain seats in the closely divided U.S. House of Representatives.