Sampling and the politics of representation in US Census 2000

Counting of people in official censuses and other social surveys produces representations that are arguably of far greater political importance than the representations produced by voting. The recent controversy around the use of sampling methods in US Census 2000 illustrates some important political - geographical dimensions of our decisions regarding whether and how to be counted in surveys. The argument is intended both to illuminate political features of this very important source of geographical data and to encourage a more self-consciously political engagement with the statisitical surveys through which modern citizens more or less consciously contribute to the shaping of our own lives.