Appendix A: Technical Notes

The State Nonfiscal Survey of Public Elementary/Secondary Education, the Local Education Agency Universe Survey, and the Public Elementary/Secondary School Universe Survey are the nonfiscal components of the Common Core of Data (CCD) survey system, while the School District Finance Survey (F-33) and the National Public Education Financial Survey (NPEFS) are the fiscal components.

Together, these five surveys provide an annual picture of the numbers of public school staff, students in prekindergarten through grade 12, and high school completers and dropouts; selected characteristics of students; numbers and types of public schools, education agencies, and programs offered; and revenues and expenditures for public elementary and secondary education. The surveys also provide directory information such as school and agency names, addresses, and telephone numbers. The data are collected from state education agencies (SEAs), processed, edited, and verified by the U.S. Bureau of the Census, the National Center for Education Statistics (NCES), and the Education Statistics Services Institute (ESSI) of the American Institutes for Research (AIR).

Data quality.

Staff at NCES, the Census Bureau, and ESSI edit all CCD data submissions and ask state CCD coordinators to correct or confirm any numbers that appear out of range when compared with other states' data or with the state's reports in previous years.

Missing data.

Not all states collect and report all of the data items requested in the CCD surveys. NCES imputes (replaces a nonresponse with an estimated value) some missing items in the state nonfiscal survey and the NPEFS. (See below for information on imputation procedures.) NCES does not impute missing items in the Public Elementary/Secondary School Universe Survey, the Local Education Agency Universe Survey, or the F-33.

When reporting results, missing data are treated differently within individual states than they are across all states, the Districts of Columbia, and other jurisdictions as a whole. When reporting data by states, an individual state is considered to have missing data if an item is reported by less than 80 percent of its schools or agencies. If a state is missing 20 percent or more of its responses for a given data item in the school or local education agency surveys, the corresponding table cell is suppressed and no count is presented. For example, in table 2 the number of magnet schools in Michigan is suppressed because the magnet school flag data item was missing for more than 20 percent of schools in the state.

When data are being reported across states, the treatment of missing data differs. When information is missing for no more than 15 percent of cases across the United States, NCES calculates totals and identifies them as totals for "reporting states," rather than for the United States, and column headings are labeled accordingly. A "reporting states" total was not calculated for the total number of magnet schools and percentage of students in magnet schools in table 2 and for the number and percentage of students receiving English language learner (ELL) services in table 3 because more than 15 percent of schools in the 50 states and the District of Columbia were missing the magnet school flag and more than 15 percent of districts in the 50 states and the District of Columbia were missing counts of students receiving ELL services. In cases where an individual state's data are suppressed for a given data item on the state level (e.g., Connecticut's number of students eligible for free or reduced-price lunch in table 3), the data reported by that state are included in the "reporting states" total (i.e., the "reporting states" total for students eligible for free or reduced-price lunch in table 3 includes students in Connecticut).

Imputed and adjusted data.

NCES has imputed and adjusted some reported values in the state nonfiscal survey and the NPEFS to create data files that more accurately reflect student, staff, and finance counts, and improve comparability among states. Imputations and adjustments were performed on data from the 50 states and District of Columbia only. It should be noted that no imputations were made to the high school graduate or other high school completer categories; however, they have been modified to prevent identification of any individual. No adjustments or imputations were performed on race/ethnicity data for any state.

Imputations provide estimates in cases in which the missing value is not reported at all, indicating that subtotals for the category are underreported. An imputation by NCES would assign a value to the missing item, and the subtotals containing this item would increase by the amount of the imputation.

Adjustments to data are corrections in cases in which a value reported for one item contains a value for one of more additional items not reported elsewhere. For example, a state may not differentiate between kindergarten teachers and prekindergarten teachers, reporting "missing" for prekindergarten teachers and a value representing the count of staff for both categories as kindergarten teachers. NCES adjusts these two responses by reducing the amount reported for kindergarten teachers and adding that amount to prekindergarten teachers.

As noted, values are imputed for some missing items in the state nonfiscal survey and the NPEFS. All imputed values in the tables are footnoted, and imputed values are never used in the imputation of another value. Missing items are imputed only for the 50 states and the District of Columbia. Totals and subtotals in tables are footnoted if one or more items in the total or subtotal are imputed or adjusted.¹

Totals.

Totals reported in the tables are limited to the 50 states and the District of Columbia. They do not include data from the Bureau of Indian Affairs, Department of Defense dependents schools (overseas and domestic), Puerto Rico, or the other jurisdictions of American Samoa, Guam, the Northern Marianas Islands, or the U.S. Virgin Islands. In cases where not all of the 50 states and the District of Columbia provided a response for a data item, but a response was provided by at least 85 percent of the schools or districts in the 50 states and the District of Columbia, a "reporting states" total is presented. See "Missing data" (above) for more information.

Membership/Enrollments.

Because some students may receive a public education outside a local school district or school (for example, they may attend a state-operated residential school), the numbers of students reported in the school or local education agency surveys are not used as the official state totals in CCD publications. The total numbers of students shown in this report are those reported in the 2004–05 State Nonfiscal Survey of Public Elementary/Secondary Education. However, the percentages of students shown in individual tables may be based on the school, local education agency, or state nonfiscal surveys, depending upon the source of the tabulated data. For example, table 2 presents the percentages of students attending charter schools, and these counts and percentages of students are based on data from the Public Elementary/Secondary School Universe Survey, where data on charter schools are collected.

School- and local education agency (LEA)-based tables in this report are limited to schools and LEAs that are operational (i.e., open) and that have membership (i.e., an enrollment of at least one student). Schools and LEAs can be reported legitimately with no student enrollment. In the CCD, schools can be reported with no student enrollment because students can be included in the enrollment for only a single school. For example, if a student is dually enrolled in a regular school and a vocational school, that student can only be reported among the membership of one of these schools. Or, a school district too small to support a school may send students to a neighboring district. In 2004–05, a total of 2,589 operational schools in the United States and 11 in the other jurisdictions reported no students in membership. Some 1,454 operational LEAs in the United States and 1 in the other jurisdictions reported no students in membership.

Student membership data from the 2003 04 State Nonfiscal Survey of Public Elementary/Secondary Education are used to calculate per pupil expenditures and revenues.

Staff counts.

All staff counts (including teachers) are reported in FTE units. This is the amount of time required to perform an assignment stated as a proportion of a full-time position. It is computed by dividing the amount of time an individual is employed by the time normally required for a full-time position.

State agencies vary in their staff data collection and reporting systems, with resulting variations across states. Several states collapsed two or more categories of staff into one. In these cases, for the state nonfiscal data only, NCES disaggregated this number by distributing the reported number of staff across the several categories based on the average distribution of these staff in the reporting states. Staff counts from the local education agency and school surveys were not disaggregated and redistributed.

High school completers.

The types of high school completion credentials available to students vary by state policy. For example, in some states the only type of completion available is a regular diploma; in others, a variety of completion categories are available, from a regular diploma to a certificate of attendance. For this reason, caution should be used when comparing state totals for diploma recipients.

"Not applicable" was accepted as a valid response for other high school completers, as some states do not offer this credential.

Confidentiality protection of high school completer data.

Under some conditions, it could be possible to identify a student who was enrolled in the 12th grade in 2003–04 and who failed to receive a regular diploma at the conclusion of that school year. For example, if a school district had one 12th-grade student in 2003–04 and no diploma recipients in 2003–04, it might be assumed that this student failed to graduate. (This would have to be an assumption, because the CCD cannot distinguish students who fail to graduate from those who transfer out of the school district.)

In order to prevent the identification of any 12th-grade student who failed to receive a regular diploma, the high school completion data were modified, either by changing some completion data to missing or by increasing the reported number of diploma recipients. These changes resulted in a minimal loss of data from the local education agency universe file.

General Educational Development (GED) test passers.

The GED Testing Service was the source for the numbers of individuals 19 or younger who passed the GED examination between October 1, 2003, and September 30, 2004. GED Testing Service data are published here because they are available for every state and are comparable from state to state. However, these numbers are approximations of the number of high school equivalencies awarded in that year. Not every individual who meets the GED Testing Service's passing score applies for and is awarded an equivalency diploma. A state may also grant an equivalency diploma on the basis of some test other than the GED, and those degrees would not be reflected in this count. Also, the state or jurisdiction in which an individual takes the GED is not necessarily the state in which he or she was last enrolled in school.

Averaged Freshman Graduation Rate (AFGR).

The AFGR provides an estimate of the percentage of high school students who graduate on time. The rate uses aggregate student enrollment data (to estimate the size of an incoming freshman class) and aggregate counts of the number of diplomas awarded 4 years later. The incoming freshman class size is estimated by summing the enrollment in 8th grade in one year, 9th grade in the next year, and 10th grade in the year after and then dividing by three. The averaging is intended to account for prior-year retentions in the 9th grade. For more information on the AFGR, please refer to Seastrom et al. (2006).

Comparability of fiscal data across states (tables 6–10).

Because the District of Columbia is a single urban district, it is often an outlier in comparisons of revenues and expenditures, with larger revenues and expenditures per student than most other school districts have. Similarly, Hawaii is a single school district and funds public education primarily through state taxes. Because of this, Hawaii's data may pose similar problems of comparability.

District-level analyses of fiscal data (table 10).

Table 10 presents revenues and expenditures per pupil. The fiscal data are taken from the School District Finance Survey (F-33), a district-level data collection for the 50 states and District of Columbia. (Tables 6–9 present data from the NPEFS, a state-level collection.) Districts included in table 10

• are listed in the Local Education Agency Universe Survey for school year 2003–04;
• have student membership greater than zero (because per pupil dollar amounts can not be calculated if a district has zero enrollment); and
• have current expenditures per pupil between $2,500 and $35,000 (in order to exclude districts with extreme values of current expenditures from the analysis; see below).

There are two scenarios in which districts can have extreme values of current expenditures: (1) when the data are correct, but the district has unusual characteristics that account for the extreme values; and (2) when the data are incorrect or were incorrectly reported on the F-33. Ideally, outliers of the first type should be included in the analysis, and those of the second type should be excluded. However, it is usually impossible to determine whether an outlier district is of the first or second type. Therefore, to exclude districts that were reported with errors, the range of current expenditures per pupil is restricted in this analysis. For more information about the selection of cutpoints for current expenditures per pupil, please refer to Hussar and Sonnenberg (2000).

Comparing expenditures across districts.

District-level analyses and comparisons can be complicated by the variety of administrative structures that exist across the nation in regular school districts. States such as Florida, Maryland, Nevada, and West Virginia have large districts that are coterminous with counties and encompass all levels and types of public schools. School districts in other states may exist in small communities with only one school, or in larger communities where all elementary schools are in one school district and all secondary schools are in another. In some states, all special education schools are administered by a few specific districts; in other states, each district may have all kinds of schools and programs. This variety in the types of school districts makes it difficult to compare expenditures across school districts. In eight states (Arizona, California, Illinois, Montana, Nebraska, New Hampshire, New Jersey, and Vermont), less than half of the school districts were unified. In two states, Montana and Vermont, less than half of the students attended schools in unified districts.

Federal range ratio.

The federal range ratio is used in this report as an indicator of the difference between districts with relatively high revenues (or expenditures) per student and districts with relatively low revenues (or expenditures) per student. As used by Berne and Stiefel (1984) and in previous NCES publications (Parish, Matsumoto, and Fowler [1995]; Hussar and Sonnenberg [2000]), the federal range ratio excludes the top and bottom 5 percent of districts in order to reduce the influence of extreme values. The federal range ratio is the difference between the amount per student of the district at the 95th percentile and the district at the 5th percentile, divided by the amount for the district at the 5th percentile.

Fiscal years.

The fiscal year begins on July 1 and ends on June 30 for most states. The fiscal year for Alabama runs from October 1 through September 30, and the fiscal year for Nebraska and Texas runs from September 1 through August 31. The F-33 and NPEFS data are not adjusted to conform to a uniform fiscal year across states.

Top

---

¹ For more information on imputation procedures, see Gray et al. (2006) and Hill and Johnson (2006).