NAEP Weighting Procedures → Weighting Procedures for the 2005 Assessment → Computation of Replicate Weights for the 2005 Assessment → Computing Student-Level Replicate Base Weights for the 2005 Assessment

## Computing Student-Level Replicate Base Weights for the 2005 Assessment

For the 2005 assessment, the calculation of the student-level replicate base weights differed by whether a student was in a school selected with certainty or not, since replicates were formed using first-stage sample units. The first-stage sample unit was the student in certainty schools and the school itself for noncertainty schools.

For students in certainty schools, where replicate strata were formed at the student-level by grouping students into pairs or triplets, student base weights were calculated for each of the 62 replicates based on this grouping. For students in replicate strata comprising pairs of students, the student-level replicate base weights STU_BWTjsk(r), r = 1,..., 62, were calculated as follows:

where

• STU_BWTjsk is the full-sample student base weight for student k in certainty school s in primary stratum j,

• Rjsr is the set of students within the r-th replicate stratum for certainty school s in primary stratum j, and

• Ujsk is the variance unit (1 or 2) for student k in certainty school s in primary stratum j.

For students in replicate strata comprising three students, two sets of student-level replicate base weights were computed (see replicate variance estimation for details): one for the first replicate r1 and another for the second replicate r2. The two sets of student-level replicate base weights STU_BWTjsk(r1), r1 = 1,..., 62, and STU_BWTjsk(r2), r2 = 1,..., 62, were calculated as described below.

For students in noncertainty schools, where replicate strata were formed at the school-level, student base weights for each of the 62 replicates STU_BWT(r), r = 1, ..., 62, were calculated as follows:

where

• SCH_BWT(r) is the replicate school base weight;

• SCHSESWT is the school-level session assignment weight used in the full-sample weight;

• WINSCHWT is the within-school student sampling weight used in the full-sample weight;

• STUSESWT is the student-level session assignment weight that reflects the conditional probability, given the particular assessment type was assigned to the school, that the student was assigned to that assessment type

• SUBJFAC is the subject factor used in the full-sample weight

• SUBADJ is the substitute adjustment factor used to account for the difference in enrollment size between the substitute and original school; and

• YRRND_AF is the year-round adjustment factor used in the full-sample weight.

Last updated 08 May 2009 (EH)