In 2006, the migration imputation procedures for the Annual Social and Economic Supplement (ASEC) of the Current Population Survey (CPS) were changed.

Prior to 2006, cases were imputed using a sort order which first sorted by the various collection subsamples that make up the full ASEC file (the regular March CPS; Hispanic households; the non-Hispanic, non-White households; and the Non-Hispanic White households with children 18 years or younger) for each of the distinct months (February, March, and April), and then within each subsample a geographic sort was applied. Beginning in the 2006 data collection cycle, data from all subsamples were geographically sorted before the implementation of the imputation algorithm, an improved method.

Research on this topic indicates that the imputation method prior to 2006 overstated interstate migration rates for imputed cases. One effect of the revised processing change was that interstate migration rates dropped significantly in 2006, but this drop is due to the improved processing method. Users of interstate migration data should be cautious of comparing rates from the 1999-2000 to 2004-5 period with other periods.