VINCENT SNOWBARGER: “I want to come at this in a little different approach. In 1992, I was the user of census products in the reapportionment…”

“I want to come at this in a little different approach. In 1992, I was the user of census products in the reapportionment in our State legislature in Kansas. We have talked about an accuracy rate back in 1990 of 98.4 percent. I think that is pretty significant. What people need to understand is that when you are using this census today to develop districts, we are looking on a block-by-block basis. We take one block, add it to another block, we aggregate those blocks together and, sooner or later, we have a Representative district or a Senate district or even a Congressional District. Right now, by the census's own numbers, the accuracy rate at the block level is plus or minus 35 percent. Thirty-five percent. It has been mentioned here several times this morning that sampling is inaccurate at the town and local level. Even the Census Bureau reports that sampling counts are less accurate than an actual head count. It is inaccurate because of this polling scheme. Small towns, including the majority of Kansas, are going to be at risk, and that is a fact. The Census Bureau's own studies prove this. The 1991 Undercount Steering Committee said, ``It is understood that for smaller areas, those with less than 100,000 population, proportionately more units would have less accurately adjusted counts than unadjusted counts.'' We just cannot use this polling method that penalizes small cities and towns.”