NAEP Technical DocumentationTrimming of School Base Weights for the 2017 Assessment

Unusually large school weights can occur under three circumstances:

  1. New Schools: When a school selected from the NAEP new-school sampling frame has an enrollment that is disproportionately large relative to the enrollment of its corresponding school district or Catholic diocese. In other words, when a large new school is selected from a small school district or Catholic diocese.
  2. Private Schools: When a school from the private school frame participates in NAEP but did not participate in the Private School Universe Survey (PSS), the source of the NAEP private school frame. Schools that fall into this category are referred to as PSS nonrespondents and have small probabilities of selection.
  3. Schools with Large Enrollment Increases: When the actual grade enrollment of a school, determined at the time of student sampling, is grossly larger than its enrollment used for school sampling.

If a school's base weight was determined to be too large, the school weight was trimmed. Recall schools were sampled for NAEP with probability proportional to size where size was based on student grade enrollment. If a sampled school had a small grade enrollment, its school base weight was large. To determine if a school's base weight was too large, a comparison was made between a school's base weight and its ideal weight (described below). If a school's base weight was more than three times its ideal weight, the school's base weight was scaled back or trimmed to three times the ideal weight. The trimming was accomplished by way of a trimming factor. The trimming factor for school s  was calculated using the following formula:


SCH underscore TRIM subscript s equals three times EXP underscore WT subscript s divided by SCH underscore BWT subscript s if open paren SCH underscore BWT subscript s divided by EXP underscore WT subscript s close paren is greater than three or equals one if otherwise

where

  • EXP_WTs is the ideal base weight for school s, and

  • SCH_BWTs is the actual school base weight for school s.

The ideal weight for a school depends on the type of circumstance, whether it was a new school, private school, or school with large grade enrollment increase. Details of the trimming procedure by type of circumstance are described below.

New Schools

New schools with a disproportionately large student enrollment in a particular grade from a school district (or Catholic diocese) that was selected with a small probability of selection were likely candidates to have their school weights trimmed. The school base weights for such schools may be large relative to what they would have been if they had been selected from the NAEP public or private school sampling frame. The ideal weight for a new school was as follows:

  • EXP_WTs is the ideal base weight the school would have received if it had been on the NAEP public or private school sampling frame.

For the 2017 NAEP assessment, six schools out of 165 participating schools selected from the new-school sampling frame had their weights trimmed: one at grade 4 and five at grade 8.

Private Schools

Private school PSS nonrespondents who participated in NAEP and were found subsequently to have either larger enrollments than assumed at the time of school sampling or an atypical probability of selection given their affiliation, the latter being unknown at the time of sampling, were also likely candidates to have their school weights trimmed. The ideal weight for a PSS nonresponding private school was as follows:

  • EXP_WTs is the ideal base weight the school would have received if it had been on the NAEP private school sampling frame with accurate enrollment and known affiliation.

For the 2017 NAEP assessment, three out of the four private school PSS nonrespondents who participated in NAEP had their weights trimmed, one at grade 4 and two at grade 8.

Schools with Large Enrollment Increases

Schools, other than the PSS nonrespondents described above, whose enrollments determined at the time of student sampling were much larger than those assumed at the time of school sampling were also candidates to have their school weights trimmed. These schools have large relative school weights because their school probabilities of selection were artificially low. The ideal weight for a school with a large grade enrollment increase was as follows:

  • EXP_WTs is the ideal base weight the school would have received if it had been on the relevant NAEP public or private school sampling frame with the updated enrollment figure from student sampling.

For the 2017 NAEP assessment, only one school with a large grade enrollment increase, a grade 8 school, had its weight trimmed.

Last updated 01 April 2022 (SK)