Table of Contents  |  Search Technical Documentation  |  References

NAEP Weighting Procedures → Weighting Procedures and Variance Estimation for the 2000 National Main Assessment → Reporting Weights Main → Poststratification Adjustment for the 2000 National Main Assessment

NAEP Technical DocumentationPoststratification Adjustment for the 2000 National Main Assessment

Poststratification is a weighting procedure that adjusts the weights of respondents so that the weighted sample distribution is the same as the known population distribution. The sums of the poststratified-adjusted weights of the respondents are equal to known population totals for certain subgroups of the population. The main purposes of poststratification are to improve precision of survey estimates by reducing their mean square error and to enhance the comparability of survey data with other surveys, particularly when comparing estimates from the same survey over time.

The poststratification adjustment procedure involves applying a ratio adjustment to the student reporting weights. Assessed and excluded students are partitioned into poststratification cells, and a single ratio adjustment factor is calculated and applied to the reporting weight of all students in a given cell. The numerator of the poststratification factor is an independent estimate of the number of students in the given cell, and the denominator is the corresponding estimate derived using the student reporting weights. The numerator is derived from 1997 and 1998 Current Population Survey (CPS) data and 1999 population projections made by the U.S. Census Bureau.

The poststratification procedure is carried out separately by grade, subject, and reporting population. Poststratification adjustment cells are defined in terms of race/ethnicity, modal age status, and Census region for all grades except grade 12. Modal age status is not used for grade 12 because it is not possible to derive reliable modal age status counts from the CPS data. CPS counts all adult education students, regardless of age, as grade 12 students. Instead, grade 12 uses poststratification cells defined by students less than or equal to modal age (i.e., seventeen years of age or younger).

The poststratification factor for student i in a given poststratification adjustment class h is given by

• TOTAL_h equals the total number of grade-eligible students in poststratification cell h from Census Bureau data;

• STU_WT_i equals the initial student weight for student i;

• STUNRADJ_i equals the student nonresponse adjustment factor for student i;

• STU_TRIM_i equals the student trimming adjustment factor for student i;

• RPT_FCTR_i equals the reporting factor for student i; and

• Set C_h equals the set of assessed and excluded students in poststratification cell h, except those at grade 12 who were age 18 or older.

Note that students at grade 12, age 18 or older, received the poststratification factor according to their poststratification cell and subject even though they were not used in calculating the factor.

Printer-friendly Version