​​​​​​​​​NAEP Technical DocumentationNew-School Sampling Frame for the 2022 Private School Long-Term Trend Assessment

The NAEP 2022 private school frame was constructed using the most current Private School Universe Survey (PSS) file available from NCES. This file contained schools that were in existence during the 2019–2020 school year, (i.e., it was two years out of date). During the subsequent 2-year period, undoubtedly, some schools closed, some changed structure (one school becoming two schools, for example), some newly opened, and still others changed their grade span.

A supplemental sample was selected from a list of Catholic schools that were new or had become newly eligible sometime after the 2019–2020 school year. The goal was to allow every new Catholic school a chance of selection, thereby fully covering the target population of Catholic schools in operation during the 2021–2022 school year. It was infeasible to ask every Catholic diocese in the United States to provide a supplemental school frame, so a two-stage procedure was employed. First, a sample of dioceses was selected. Then the National Catholic Educational Association (NCEA) was sent a list of the schools within their sampled dioceses that had been present on the 2019–2020 PSS file. NCEA was asked to add in any new schools and update grade span for the schools on this list.

The new-school process began with the preparation of a diocese-level frame. The starting point was a file containing every Catholic diocese in the U.S. classified as small, medium, or large based on the number of schools and student enrollment of schools from the PSS private school frame. The new-school process for long-term trend (LTT) piggybacked on the process for the grade-based samples as follows:

A diocese was considered to be small if it contained no more than one school at each of grades 4 and 8. During school recruitment, schools sampled from small dioceses were asked to identify schools within their dioceses that newly offered the targeted grades (grades 2-5 for age 9). From a sampling perspective, each new school was viewed as an "annex" to the sampled school, which meant that it had a well-defined probability of selection equal to that of the sampled school. When a school in a small diocese was sampled from the PSS frame, its associated new school was automatically sampled as well.

Dioceses that were not small were further divided into two strata, one containing large-size dioceses and a second containing medium-size dioceses. These strata were defined by computing the percentage of grade 4 and 8 enrollment represented by each diocese, sorting in descending order, and cumulating the percentages. All dioceses up to and including the first diocese at or above the 80th cumulative percentage were defined as large dioceses. The remaining dioceses were defined as medium dioceses.

A simplified example is given below. The dioceses are ordered by descending percentage enrollment. The first six become large dioceses and the last six become medium dioceses.

In actuality, there were 77 large and 96 medium dioceses in the sampling frame.

The target sample size was 10 dioceses total across the medium-size and large-size diocese strata: eight from the large-size diocese stratum and two from the medium-size diocese stratum.

In the medium-size diocese stratum, dioceses were selected with equal probability. In the large-size diocese stratum, dioceses were sampled with probability proportional to enrollment. These probabilities were retained and used in later stages of sampling and weighting of new schools.

NCEA was sent a listing of all the schools in the selected dioceses that appeared on the 2019–2020 PSS file and was asked to provide information about the new schools not included in the file and grade span changes of existing schools. These listings were used as sampling frames for selection of new Catholic schools and updates of existing schools, keeping in mind that grades 2-5 were targeted for age 9. In addition, the new-school frames were limited to the geographic areas covered by the sampled LTT primary sampling units (PSUs).​

The following table presents the number and percentage of schools and average estimated age enrollment for the age 9 "new-school" frames by census region. For age 9 there were no new schools in the Midwest region. 

Number and percentage of schools and mean school size in the private new-school frame, long-term trend assessment for age 9, by census region: 2022

​Census region	Schools	Percentage	Mean school size
​Total	15	100.00	32
Northeast	11	73.33	37
Midwest	0	0.00	0
South	3	20.00	11
West	1	6.67	35
SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2022 Long-Term Trend Mathematics and Reading Assessments.

​​For age 13 a private new-school frame was not constructed. Based on the 2022 experience with age 9, where only one of the new schools on the frame was sampled, it was decided not to conduct a private new-school procedure for age 13. Because the age 13 assessment was in a different school year than age 9, conducting the new-school procedure for age 13 would have required that the process described above be carried out anew, one year after the age 9 process was conducted.
​

Printer-friendly Version