Table of Contents  |  Search Technical Documentation  |  References

NAEP Weighting Procedures → 2002 Weighting Procedures and Variance Estimation → National Main Assessment Weights → Computation of Base Weights → Computation of Base Weights for Originally Sampled Schools

NAEP Technical DocumentationComputation of Base Weights for Originally Sampled Schools

The starting point for any school base weight is the inverse of the school probability of selection . In the 2002 NAEP assessment, there was also a session assignment factor, SCHSES_s^(A), that reflected the probability of assigning at least one "A" session to the school. (An "A" session is a session with reading and writing operational assessments.) These session assignment factors vary by grade, school type (public/private), and the size of the school. The session assignment factor is 1 for public schools in fourth and eighth grade, twelfth-grade public schools with at least 35 students, eighth-grade private schools with at least 24 students, and twelfth-grade private schools with at least 24 students. All of these schools had at least one "A" session. The session assignment factor is equal to 1.43 (10/7) for twelfth-grade schools with less than 35 students (as these schools had a 70 percent chance of being assigned at least one "A" session). Similarly, for fourth-grade private schools, and eighth-grade and twelfth-grade private schools with less than 24 students, the session assignment factor is equal to the reciprocal of the probability that the private school is assigned at least one "A" session.

The initial school weights for assignment into Session "A" can be written as w_s ×  SCHSES_s^(A) 

where

• w_s is the inverse of the school probability of selection; and

• SCHSES_s^(A) is the session allocation weight, the inverse of the probability that the school is assigned at least one "A" session.

It is important to note that w_s for the twelfth-grade public schools is equal to the unconditional probability of selection, before adjustments for overlap control with the Educational Longitudinal School (ELS) survey.

The NAEP sampling and weighting staff compared the estimated school enrollment by grade and jurisdiction, derived from the sample, using base weights, with frame estimates. Since the full school sample weighted enrollments are unbiased estimates of the frame values, the difference between the sample weighted enrollment and the frame enrollment for each jurisdiction should reflect the sampling error. The webpage Weighted Enrollment from the 2002 National Main School Samples provides documentation of the checks made between the weighted school sample and the frame.

Printer-friendly Version