This is related to this question.
I understand that the recommendation in this question was to use the ACS weights even when dealing with a subnational analysis. However, I would like some precision on the weighting design of the 1990 and 2000 censuses. We are constructing a panel of MSA-level statistics using the 1970, 1980, 1990, and 2000 censuses, and the ACS of 2007. What we are most concerned about is representativeness at the MSA-level and comparability across time.
According to the Sample Design page, “The [weighted] sample is not only more cumbersome to use than those previously produced by the Census Bureau, but precision is actually reduced for the general population.”
Therefore, should I use the 1% unweighted samples of the 1990 and 2000 censuses instead of the 5% and not care about the weights, or should I use the weights in the 5% samples?