The procedures for generating full sample weights and replicate weights for the student samples for the geography and U.S. history assessments include:
base weights reflecting school and student sampling,
base weight factors reflecting assignment to field sessions and assignment to U.S. history or geography sessions,
nonresponse adjustments for school and student nonresponse,
trimming adjustments to reduce variability in the sample weights, which reduces mean square error,
poststratification adjustments to the Census Bureau's Current Population Survey (CPS) control totals (also to reduce mean square error), and
replicate weights for the computation of jackknife variance estimates.
The sample design that determines the base weights is discussed in Sample Design for the 2001 Assessment.