Appendix B—Methodology and Technical Notes—Sampling Frames and Sample Selection

Public and BIE schools. The starting point for the 2007-08 SASS public school and BIE school sampling frame was the preliminary 2005–06 Common Core of Data (CCD) Nonfiscal School Universe data file. The CCD includes regular and nonregular schools (special education, alternative, vocational, or technical), public charter schools, and BIE schools. The sampling frame was adjusted from the CCD in order to fit the definition of a school eligible for SASS. To be eligible for SASS, a school was defined as an institution or part of an institution that provides classroom instruction to students; has one or more teachers to provide instruction; serves students in one or more of grades 1–12 or the ungraded equivalent; and is located in one or more buildings apart from a private home. It was possible for two or more schools to share the same building; in this case, they were treated as different schools if they had different administrators (i.e., principal or school head).

The SASS definition of a school is generally similar to the CCD definition, with some exceptions. SASS is confined to the 50 states plus the District of Columbia and excludes the other jurisdictions and Department of Defense overseas schools. The CCD includes some schools that do not offer teacher-provided classroom instruction in grades 1–12 or the ungraded equivalent. In some instances, schools in the CCD are essentially administrative units that may oversee entities that provide classroom instruction or they may only provide funding and oversight. The CCD schools with the same location, address, and phone number were collapsed during the SASS frame building on the assumption that the respondent would consider them to be one school. Because SASS allows schools to define themselves, Census Bureau staff observed that schools generally report as one entity in situations where the administration of two or more schools on the CCD is the same. A set of rules was applied in certain states to determine in which instances school records should be collapsed together. When school records were collapsed together, the student and teacher counts, grade ranges, and names as reported to the CCD were all modified to reflect the change.

Finally, additional school records were added to the sampling frame. Most of these records were for Career Technical Centers or alternative, special education, or juvenile justice facilities in California, Pennsylvania, New York, and other states. For a detailed list of frame modifications, see the Documentation for the 2007–08 Schools and Staffing Survey (Tourkin et al. forthcoming). After adding, deleting, and collapsing school records, the SASS public school sampling frame consisted of 90,410 traditional public schools, 3,850 public charter schools, and 180 BIE schools.

The SASS sample is a stratified probability-proportionate-to-size (PPS) sample. With the exception of BIE schools, all schools underwent multiple levels of stratification.³ The sample was allocated so that national-, regional-, and state-level elementary, secondary, and combined public school estimates could be made. The sample was allocated to each state by grade range (elementary, secondary, and combined) and school type (traditional public, public charter, BIE-funded, and schools with high American Indian enrollment). For a full description of the allocation procedure, see the Documentation for the 2007–08 Schools and Staffing Survey (Tourkin et al. forthcoming). Within each stratum, all non-BIE schools were systematically selected using a PPS algorithm. The measure of size used for the schools was the square root of the number of full-time equivalent teachers reported or imputed for each school during the sampling frame creation. Any school with a measure of size greater than the sampling interval (the inverse of the rate at which the sample is selected) was included in the sample with certainty and thus automatically excluded from the probability sampling operation. (For a more detailed explanation of PPS sampling, consult Sampling Techniques [Cochran 1977].) These sampling procedures resulted in a total public school sample of about 9,800 public schools (including both traditional public and public charter schools) and 180 BIE schools in the 2007–08 SASS.

Private schools. The 2007–08 SASS private school frame was based on the 2005–06 Private School Universe Survey (PSS).⁴ The PSS is designed to generate biennial data on the total number of private schools, students, and teachers and to build a universe of private schools in the 50 states and the District of Columbia to serve as a sampling frame of private schools for NCES sample surveys, such as SASS. Prior to each PSS collection, NCES updates the list of schools to be included in the PSS by collecting membership lists from private school associations and religious denominations, as well as private school lists from state education departments. This list-frame update is intended to improve coverage of new private schools and private schools previously not included in the PSS. Since the list-frame update for the 2007–08 PSS preceded the SASS frame building, the SASS private school sample frame was augmented with the private schools discovered during the PSS list-frame update. The SASS private school frame also includes schools in the 2005–06 PSS area frame. This area frame includes schools that were overlooked in the list-building operation. In a sample of 124 geographical areas (out of 2,062), local field staff used such resources as local telephone directories, local resource guides, etc., to identify schools. The resultant lists of schools were matched to the PSS list frame, and any school that was not part of the list frame was added as an area frame school.

During the creation of the SASS private school sampling frame, schools with kindergarten as the highest grade level were deleted from the frame. Also, private school records that were missing information necessary for the SASS school sample selection were amended. For example, the school grade range was needed to stratify SASS schools during the private school sampling process. If the data on school grade range were missing, values were assigned in one of four ways: taking information from earlier PSS data, using information from the school's name (e.g., middle school) to assign a grade range, calling the school to assign a specific grade range, or, as a last resort, assigning a grade level of combined (both elementary and secondary levels). Missing information on school affiliation and student and teacher counts were assigned in a similar manner. After these changes, the private school sampling frame consisted of 28,490 private schools.

Private schools were stratified by affiliation, grade level (elementary, secondary, and combined), and census region (Northeast, Midwest, South, and West). The number of private school affiliation strata was reduced from 17 in 2003–04 to 11 in 2007–08. The current affiliation strata include

• Catholic – parochial;
• Catholic – diocesan;
• Catholic – private;
• Baptist;
• Jewish;
• Lutheran;
• Seventh-Day Adventist;
• Other religious;
• Nonsectarian – regular;
• Nonsectarian – special emphasis; and
• Nonsectarian – special education.

Within each stratum, private schools in the list frame were systematically selected using a PPS algorithm. The measure of size used was the square root of the number of full-time equivalent teachers. Any school with a measure of size larger than the sampling interval was excluded from the probability sampling process and included in the sample with certainty. About 2,940 private schools were sampled, 2,760 from the list frame and 180 from the 2005–06 PSS area frame.

School districts. Since the SASS sample design calls for schools to be selected first, the school district sample consists of the districts that were associated with the schools in the public school sample. This provides the linkage between the district and the school. However, in Delaware, Florida, Maryland, Nevada, and West Virginia, a different sampling method was used because an earlier simulation study revealed that in these states standard errors were high relative to the sampling rate. To improve the reliability of SASS school district estimates, all districts in these states were included in the sample. Placing all districts in each of these five states in the sample reduced the overall state standard error to zero, if all districts responded. About 5,250 public school districts were pulled into the sample by being associated with sampled public schools.

Teachers. Teachers are defined as staff who teach regularly scheduled classes to students in any of grades K–12. Teacher rosters (i.e., Teacher Listing Forms) were collected from sampled schools, primarily by mail, and compiled at the Census Bureau. This compilation was done on an ongoing basis throughout the roster collection period. Along with the names of teachers, sampled schools were asked to provide information about each teacher's teaching experience (1–3 years, 4–19 years, and 20 or more years), teaching status (full- or part-time), and subject matter taught (special education, general elementary, math, science, English/language arts, social studies, vocational/technical, or other), as well as whether the school felt the teacher would likely be teaching at the same school the following year.

Sampling was also done on an ongoing basis throughout the roster collection period. The Census Bureau first stratified teachers into five teacher types: (1) new teachers expected to be teaching at the same school the next school year, (2) experienced teachers expected to be teaching at the same school the next school year, (3) new teachers expected to leave the next school year, (4) mid-career teachers expected to leave the next school year, and (5) highly experienced teachers expected to leave the next school year. Before teachers were allocated to these strata, schools were first allocated an overall number of teachers to be selected within each school stratum. Sampling rates for teachers varied between the strata listed above. Private school teachers in above category 2 and all teachers in above categories 3–5 were oversampled at different rates. So that a school would not be overburdened by sampling too large a proportion of its teachers, the maximum number of teachers per school was set at 20. About 14 percent of the eligible public schools, 15 percent of the eligible private schools, and 13 percent of eligible BIE schools did not provide teacher lists. For these schools, no teachers were selected. Within each teacher stratum in each school, teachers were selected systematically with equal probability. About 56,360 teachers were sampled, 47,440 from public schools, 750 from BIE schools, and 8,180 from private schools (detail may not sum to totals because of rounding).

Principals. The principal of each sampled school was selected. About 12,910 school principals were sampled (9,800 public, 180 BIE, and 2,940 private) (detail may not sum to totals because of rounding).

Library media centers. A library media center in each sampled public and BIE school was also selected. Private schools were not included in the library media center survey because of NCES budget constraints. About 9,800 public school and 180 BIE school library media centers were sampled.

Top

---

³ BIE schools that fit the SASS definition of a school were included in the 2007–08 SASS public school sample with certainty.
⁴ For more information about the PSS, see PSS website.