@article{WACHTER:2000:MLHWCD,
   AUTHOR = {Wachter, Kenneth W. and Freedman, David A.},
    TITLE = {Measuring local heterogeneity with 1990 census data},
  JOURNAL = {Demographic Research},
   VOLUME = {3},
   NUMBER = {10},
     YEAR = {2000},
      URL = {http://www.demographic-research.org/volumes/vol3/10/},
 ABSTRACT = {A sample covering 204,394 blocks from the 1990 U.S. Census permits
             measurement of residual heterogeneity from local area to local area
             after controlling by stratification for demographic characteristics
             such as race, ethnicity, age, sex as well as geographic
             characteristics such as region and place-type. The local areas have
             populations on the order of 10,000 people. The variables studied
             are four indices of enumeration difficulty. The results show that
             variance due to heterogeneity from area to area is comparable to
             (if not larger than) variance from stratum to stratum and can be
             expected to dominate sampling variance---especially with samples as
             large as the ones used in the U.S. Census Bureau's Post-Enumeration
             Surveys. These findings constrain the viable estimation strategies
             that could be employed for local tallies in the U.S. 2000 Census.},
   }

        

@article{WACHTER:2000:TFCCBI,
   AUTHOR = {Wachter, Kenneth W. and Freedman, David A.},
    TITLE = {The fifth cell: Correlation bias in U.S. census adjustment},
  JOURNAL = {Evaluation Review},
   VOLUME = {24},
     YEAR = {2000},
    PAGES = {191--212},
      URL = {http://erx.sagepub.com/cgi/content/abstract/24/2/191},
 ABSTRACT = {One form of error that can affect census adjustments is correlation
             bias, reflecting people who are doubly missing--from the census and
             from the adjusted counts as well. This article presents a method
             for estimating the total national number of doubly-missing people
             and their distribution by race and sex. Application to the 1990
             U.S. census adjustment leads to an estimate of 3 million doubly-
             missing people. Correlation bias is likely to be a serious problem
             for census adjustment in 2000. The methods of this article are well
             suited for measuring its magnitude.},
   }