Table of Contents  |  Search Technical Documentation  |  References

NAEP Sample Design → NAEP 2003 Sample Design → 2003 National Main Assessment Sample Design → Sampling of Schools for the 2003 National Main Assessment → Private School Selection for the 2003 National Main Assessment

NAEP Technical DocumentationPrivate School Selection for the 2003 National Main Assessment

The private school sample was drawn to cover both the national main mathematics and reading assessments and the pilot tests. This discussion focuses on the national main assessment sampling process, since the pilot test samples were embedded in the operational assessment samples for fourth and eighth grades. The twelfth-grade sample was limited to pilot testing only.

The table below indicates the target private school student sample sizes for the 2003 national main assessment. The third column in the table indicates the target student assessment yields for the national main mathematics and reading assessments. These overall targets were adjusted upwards to allow for school and student attrition from nonresponse, based on the observed attrition rates from the 2002 assessment (for each grade and private school subgroup separately). The final targets in the last column of the table were obtained by further inflation of the adjusted targets by five percent to account for ineligibility of schools, exclusion of students, and other attrition factors. The final targets in the last column were then used to determine school sample sizes and sampling rates.

Target private school student sample sizes, by grade, national main assessment, by private school stratum: 2003

Grade	Private school stratum	National main assessment target	NAEP 2002 school yield rate	NAEP 2002 student yield rate	Attrition-adjusted target
All	All private	25,800	†	†	40,252
4	Total private	12,600	†	†	18,531
	Catholic	6,300	0.90	0.94	7,820
	Lutheran	1,575	0.86	0.93	2,068
	Conservative Christian	1,575	0.68	0.93	2,615
	Other private	3,150	0.59	0.93	6,028
8	Total private	12,600	†	†	20,363
	Catholic	6,300	0.87	0.94	8,089
	Lutheran	1,575	0.83	0.94	2120
	Conservative Christian	1,575	0.60	0.93	2,964
	Other private	3,150	0.50	0.92	7,190
12	Total private	600	†	†	1,358
	Catholic	300	0.69	0.87	525
	Other private	300	0.45	0.84	833
† Not applicable; totals are calculated for counts only. Note: The 12th grade private school stratification was limited to Catholic and all other private schools(combined into a single group) because oversampling of private schools other than Catholic was not required in the same degree of detail for 12th grade in the 2003 assessment. In addition, 12th grade schools were included for pilot testing only. SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2003.

The private school sampling frame also included schools without a known affiliation (Catholic, Lutheran, Conservative Christian, or Other Private). These schools were sampled in a separate stratum to make sure that the private school samples were fully representative. The target for this group was set at 25 schools for each grade.

The procedure for setting measures of size to achieve the student sample targets was essentially the same procedure as for public schools, with each private school stratum done separately (including the affiliation-unknown stratum described in the preceding paragraph). The initial measure of size MOS_js for school s in each private school stratum j is set as follows as a function of x_js, the total number of students in school js in the fourth (eighth, twelfth) grade:

where w_js is the Private School Survey school base weight.

Multiple hits were allowed in the eighth- and twelfth-grade samples. The upper bounds u_j on hits were 1 for fourth-grade private schools, 2 for eighth-grade private schools, and 3 for twelfth-grade private schools. The final b_j values following the iterative process are given in the following table.

Sampling parameter values b_j for private schools, by grade, national main assessment, by private school stratum: 2003

Private school stratum	Fourth grade bj value	Eighth grade bj value	Twelfth grade bj value
Roman Catholic	0.000625329	0.000762705	0.000006310
Lutheran	0.001823661	0.002518426	0.000119034
Conservative Christian	0.000817560	0.001182344	0.000119034
Other private	0.001064906	0.001344581	0.000119034
Affiliation unknown	0.000280027	0.000280027	0.000119034
SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2003.

The measures of size E_js were calculated then as follows:

These measures were the final measures of size for the fourth- eighth- and twelfth-grade samples.

Printer-friendly Version