Census should not ignore undercounted

How many Americans are there? It depends on who’s counting and — increasingly — what their political persuasion is.

That’s the unmistakable conclusion one reaches with word that the Bush administration is likely to formally accept Census data that seriously undercounts broad sections of American society.

The motivation is political, but the results will damage the opportunities of some areas of the country and some groups of people to obtain federal help.

The issue is what set of numbers will be used in the final, official tally: Will it be "raw" numbers, those actually counted, or will the raw numbers be adjusted through scientifically accepted sampling techniques?

The professionals at the Census Bureau prefer the adjusted figures, which add in about 4 million undercounted people. But Bush Commerce Secretary Don Evans said he, not the Census Bureau, will make the final decision on what set of numbers to accept.

It is widely believed Evans will chose the lower, door-to-door count.

The politics in this is that the adjusted figures tend to add population in areas where Democrats do well, among lower-income and immigrant groups, for instance.

That may help explain why former Commerce Secretary William Daley wanted the Census Bureau to make the decision. He had science on his side, but he also knew that the decision would be likely to benefit Democrats.

The U.S. Supreme Court has already said that reapportionment (dividing up the country for congressional seats) should use the raw data. So there is no good political reason not to use the more accurate data for followup federal programs, ranging from welfare aid to federal support for highways and schools.

The point of the census is to find out how many Americans there are and where they live. There’s no reason not to make this process as accurate as possible.

[back to top]