A new feature of the 2003 State NAEP design was the integration of the Trial Urban District Assessment (TUDA) samples into the State design from the beginning of the design process. The TUDA samples were introduced in 2002 as an experiment to determine the feasibility of conducting district-level NAEP assessments.
The selection of schools for the TUDA samples followed the same design as the jurisdiction samples for State NAEP and was integrated into the state assessment design. The sample size requirements for the states and TUDA districts were not additive, due to overlapping, since each district lies within a state. For states with TUDA districts, sampling rates set for the part of the state that contained the district differed from the rates set for the rest of the state.
The procedure for defining school measures of size based on the TUDA student target sample sizes, and based on the estimated grade enrollments for the schools within these districts, was identical to that defined for the main samples. The maximum number of hits was 1 for all districts for the fourth-grade sample, and for Chicago, Houston, Los Angeles, New York City and San Diego in the eighth-grade sample. Atlanta had a maximum number of 4 hits for eighth grade, and Albuquerque, Boston, Charlotte and Cleveland had a maximum number of 3 hits for eighth grade.
For the eighth grade sample, the school probability of selection (πjs) was 1 in several districts. For the remaining districts, preliminary probabilities of selection πjs for each school were set equal to
Note that the measure of size (MOS) formula is the same as used for the initial school sample. Because of the overlap with the State NAEP jurisdiction samples for the states with TUDA districts, schools which were located in the TUDA districts were sampled at the rate called for by the TUDA design parameters. Schools not in the TUDA districts were sampled at the state rate. Thus, sample sizes in the affected states were somewhat higher than the targets set for the state sample.