The total population calculated from the 2000 census 5% sample (using perwt) is 5.6e+08, which is almost twice as large as the actual total U.S. population in 2000.
Is there an explanation for this discrepency? The total population calculated from other years all seem quite reasonable.
Is there any chance you have altered PERWT in 2000 or manipulated the data for 2000 in any other way? I just created an extract and checked the values and there doesn’t seem to be an issue. I would suggest re-saving your extract and trying the tabulation again.