NCES Handbook of Survey Methods

Schools and Staffing Survey (SASS)

Website: http://nces.ed.gov/surveys/sass/
Updated: August 2018

1. OVERVIEW SAMPLE SURVEY OF

PUBLIC, PRIVATE,
The NCES Schools and Staffing Survey (SASS) provides data on public and private
CHARTER, AND BIE
schools, principals, school districts, and teachers. SASS gathers information about many
SCHOOLS
topics, including various characteristics of elementary and secondary students, some of
the professional and paraprofessional staff who serve them, the programs offered by SASS collects data on:
schools, principals’ and teachers’ perceptions of school climate and problems in their
schools, teacher compensation, and district hiring practices. SASS is a unified set of  School districts
surveys that facilitates comparisons between public and private schools and allows
linkages of teacher, school, school district, principal, and library media center data. First  Principals
conducted in school year 1987–88, SASS has been conducted seven times, most recently
in school year 2011–12.  Schools

Purpose  Teachers
The purpose of SASS is to collect the information necessary for a complete picture of
American elementary and secondary education. SASS is designed to provide national  Library media centers
estimates of public elementary, secondary, and combined schools and teachers; state
estimates of public elementary and secondary schools and teachers; and estimates for
private schools; teachers and principals at the national level; and by private school
affiliation. The SASS questionnaires were revised for the 2003–04 and the 2007–08
administrations, with the addition of new items about teachers’ career paths, parental
involvement, school safety, and institutional support for information literacy. The
questionnaires continued to measure the same five policy issues: teacher shortage and
demand; characteristics of elementary and secondary teachers; teacher workplace
conditions; characteristics of school principals; and school programs and policies.

Core Components
SASS consists of four core components administered to districts, schools, principals, and
teachers. The district questionnaire is sent to a sample of public school districts. The
school questionnaire is sent to a sample of public schools and private schools, as well as
all charter schools in operation as of 1998–99, and all schools operated by the Bureau of
Indian Education (BIE) or American Indian/Alaska Native tribes. However, the BIE
collection was discontinued after the 2007–08 administration. The principal and teacher
questionnaires are sent to a sample of principals and teachers working at the schools that
receive the school questionnaire. There are two follow-up surveys: the Teacher Follow-
up Survey (which is covered in the TFS chapter) and the Principal Follow-up Survey
(which is covered in the PFS chapter).

SASS-1
 NCES Handbook of Survey Methods

School District Survey (formerly the Teacher Demand and school performance, and time allocation for students
and Shortage Survey). The questionnaire for this survey is during the week were added or revised in the 2007–08
mailed to each sampled local education agency (LEA). The questionnaire. These changes were then carried through to
respondents are contact people identified by LEA the 2011–12 questionnaire.
personnel.
School Survey. The questionnaires for this survey are sent
If no contact person is identified, the questionnaire is to public schools, private schools, BIE schools 1, and
addressed to “Research Director.” The School District charter schools. Private schools receive the Private School
Questionnaire consists of items about student enrollment, Questionnaire, while BIE schools and charter schools
number of teachers, teacher recruitment and hiring receive the School Questionnaire (with district items),
practices, teacher dismissals, existence of a teacher union, described separately below. As in 2007–08, the 2011–12
length of the contract year, teacher compensation, school data collection for the private school component of SASS
choice, magnet programs, graduation requirements, coincided with the administration of the NCES Private
oversight of home-schooled students and charter schools, School Universe Survey (PSS). Since both PSS and SASS
use of school performance reports, migrant education, and were administered in 2011–12, to reduce respondent
professional development for teachers and administrators. burden, the private schools in the SASS sample were not
Some items that appeared previously have been dropped, sent a PSS questionnaire. Instead, the PSS items appeared
such as those that collected layoff data and counts of in the SASS Private School Questionnaire. (See the PSS
students by grade level (the latter are available through the chapter for a more thorough discussion.)
NCES Common Core of Data [CCD]). In the 2003–04
administration, new topics, including principal hiring The School Questionnaire is addressed to “Principal,”
practices and instructional aide hiring practices, were although the respondent can be any knowledgeable school
added to the questionnaire. In the 2007–08 administration, staff member (e.g., vice principal, head teacher, or school
items on district performance, teacher tenure and administrator). Items cover grades offered, student
dismissal, principal salary, length of the contract year for attendance and enrollment, staffing patterns, teaching
teachers, and type of retirement benefits for teachers were vacancies, high school graduation rates, programs and
added or revised. services offered, curriculum, and college application rates.
The Private School Questionnaire also includes items from
The School District Questionnaire is mailed only to public the School District Questionnaire that are applicable to
school districts. Independent public charter schools, BIE- private schools. The 2007–08 collection included items on
funded schools1, and schools that are the only school in the the beginning time of students’ school day; length of the
district are given the School Questionnaire (with district school year for students; school websites; and math,
items), not the School District Questionnaire. The School reading, or science specialist assignments.
Questionnaire (with district items) includes all of the items
included in the School Questionnaire as well as selected School Questionnaire (with district items). The purpose of
items from the School District Questionnaire. The the questionnaire (which was also referred to as the
applicable items for private schools appear in the Private Unified School Questionnaire in the 2003–04 SASS) was
School Questionnaire. to obtain information about schools, such as grades
offered, number of students enrolled, staffing patterns,
School Principal Survey (formerly the School teaching vacancies, high school graduation rates, programs
Administrator Survey). The questionnaire for this survey and services offered, and college application rates. Schools
collects information about principal/school head that are the only school in the district, state-run schools
demographic characteristics, training, experience, salary, (e.g., schools for the blind), charter schools that do not
and judgments about the seriousness of school problems. report to a traditional school district, and BIE-funded
Information is also obtained on professional development schools1 received the School Questionnaire (with district
opportunities for teachers and principals, teacher items), an expanded version of the Public School
performance, barriers to dismissal of underperforming Questionnaire that included items from the School District
teachers, school climate and safety, parent/guardian Questionnaire.
participation in school events, and attitudes about
educational goals and school governance. The 2007–08 Teacher Survey. The questionnaire for this survey is
questionnaire appeared in two versions: one for principals mailed to a sample of teachers from the SASS sample of
or heads of public schools and one for heads of private
schools. The two versions contain minor variations in
phrasing to reflect differences between public and private
1The BIE data collection was discontinued after the 2007-08
SASS; therefore no BIE schools, principals, teachers or library
schools in governing bodies and position titles in schools.
media centers were sampled for the 2011-12 SASS.
Items on experience prior to becoming a principal, teacher

SASS-2
 NCES Handbook of Survey Methods

schools. It is sent to teachers in public schools, private the 2003–04 questionnaire. Items on access to online
schools, charter schools, and BIE schools. The Teacher licensed databases, resource availability, and information
Questionnaire collects data from teachers about their literacy were added or revised in the 2007–08
education and training, teaching assignment, certification, questionnaire.
workload, and perceptions and attitudes about teaching.
Questions are also asked about teacher preparation, School Library Media Specialist/Librarian Survey. The
induction, organization of classes, computers, and questionnaire for this survey was mailed to a subsample of
professional development. The only eligible respondent for the SASS sample of public, private, and BIE schools in
each teacher questionnaire is the teacher named on the 1993–94. The survey solicited data that could be used to
questionnaire label. As of the 1993–94 SASS, describe school librarians—for example, their educational
administrators are eligible for both the Teacher Survey and background, work experience, and demographic
the Principal Survey, if they teach a regularly scheduled characteristics. Because much of the collected information
class. In the 2007–08 Teacher Survey, items on grade was comparable to that obtained in the Teacher
range of teaching certification, use of electronic Questionnaire, comparisons between librarians and
communications with parents, and out-of-pocket expenses classroom teachers can be made.
for school supplies were added or revised.
Periodicity
Teacher Listing Form. The SASS Teacher Listing Form Between the 1987–88 and 1993–94 school years, SASS
collects the full list of teachers from a school, along with core components were on a 3-year cycle, with the TFS
information on subject matter taught, full- or part-time conducted 1 year after SASS. After a 6-year hiatus, SASS
teaching status, and teaching experience. The information was fielded again in the 1999–2000, 2003–04, 2007–08,
in the Teacher Listing Form is used to select a school years and most recently the 2011–12 school year
representative teacher sample and send out the Teacher (with the TFS following in 2000–01, 2004–05, 2008–09,
Questionnaires. In 2007–08, the Teacher Listing Form and 2012–13). Since 1999–2000, SASS administrations
restored a section that was removed in 2003–04, which had have been scheduled on a 4-year cycle.
asked about the school name and grade range for
verification purposes. (This section was not included in the Data Availability
survey questionnaire in 2003–04, as it was verified at the Information on all years of SASS data is available at
school, using a laptop-collected form.) https://nces.ed.gov/surveys/sass/dataproducts.asp.

Additional Components 2. USES OF DATA

In addition to the core data collection described above,
SASS featured additional components focusing on library SASS is the largest, most extensive survey of school
media specialists/librarians and on student records in districts, schools, principals, teachers, and library media
1993–94 and on library media centers in 1993–94, 1999– centers in the United States today. It includes data from the
2000, 2003–04, 2007–08, and 2011–12. One year public, private, and BIE school sectors. Moreover, SASS is
following each SASS, a Teacher Follow-up Survey (TFS) the only survey that studies the complete universe of
is mailed to a sample of participants in the SASS Teacher public charter schools. Therefore, SASS provides a
Survey. (See the TFS chapter for a more complete multitude of opportunities for analysis and reporting on
description.) As well, one year following the the 2007–08 issues related to elementary and secondary schools.
and 2011–12 iterations of SASS, a Principal Follow-up
Survey occurred (2012–13 PFS); SASS data have been collected seven times between 1987
(http://nces.ed.gov/surveys/sass/question1213pfs.asp). and 2011. Many questions have been asked of respondents
at multiple time points, allowing researchers to examine
School Library Media Center Survey. This survey was trends on these topics over time. SASS asks similar
added in the 1993–94 SASS. The questionnaire for the questions of respondents across sectors, including public,
survey asks public and BIE schools about their access to public charter, BIE, and private schools. The consistency
and use of new information technologies. The of questions across sectors and the large sample sizes
questionnaire was not sent to private schools in 2003–04, allow for exploration of similarities and differences across
due to budgetary reasons. (In 2007–08 and 2011–12, the sectors.
survey only surveyed public schools as well.) The survey
collects data on library collections, media equipment, use SASS data are representative at the state level for public
of technology, staffing, student services, expenditures, school respondents and at the private school affiliation
currency of the library collection, and collaboration level for private school respondents. Thus, SASS is
between the library media specialist and classroom invaluable for analysts interested in elementary, middle,
teachers. A section on information literacy was added to and secondary schools within or across specific states or
private school affiliations. The large SASS sample sizes

SASS-3
 NCES Handbook of Survey Methods

allow extensive disaggregation of data according to the cooperatives that employ special education teachers who
characteristics of teachers, administrators, schools, and teach in schools in more than one school district.)
school districts. For example, researchers can compare
urban and rural settings and the working conditions of Public School. An institution that provides educational
teachers and administrators of differing demographic services for at least one of grades 1–12 (or comparable
backgrounds. ungraded levels), has one or more teachers to give
instruction, is located in one or more buildings, receives
SASS collects extensive data on teachers, principals, public funds as primary support, and is operated by an
schools, and school districts. Information on teachers education agency. Schools in juvenile detention centers
includes their qualifications, early teaching experience, and schools located on military bases and operated by the
teaching assignments, professional development, and Department of Defense are included.
attitudes about the school. The SASS School Principal
Questionnaire collects information about principals’ or Private School. An institution that is not in the public
school heads’ years of experience and training, goals and system and that provides instruction for any of grades 1–12
decision making, professional development for teachers (or comparable ungraded levels). The instruction must be
and instructional aides, school climate and safety, student given in a building that is not used primarily as a private
instructional time, principal perceptions and working home. Private schools are divided into three categories: (1)
conditions, and demographic information. Questions about Catholic: parochial, diocesan, private order; (2) other
schools include enrollment, staffing, the types of programs religious: affiliated with a conservative Christian school
and services offered, school leadership, parental association, affiliated with a national denomination,
involvement, and school climate. At the district level, unaffiliated; and (3) nonsectarian: regular, special program
information is sought on the recruitment and hiring of emphasis, special education. The classification of
teachers, professional development programs, student nonsectarian schools by program emphasis disentangles
services, and other relevant topics. private schools offering a conventional academic program
(regular) from those that either serve special-needs
SASS data can be very useful for researchers performing children (special education) or provide a program with a
their own focused studies on smaller populations of special emphasis (e.g., arts and sciences).
teachers, administrators, schools, or school districts. SASS
can supply data at the state, affiliation, or national level Charter School. A charter school is a public school that, in
that provide valuable contextual information for localized accordance with an enabling state statute, has been granted
studies; localized studies can provide illustrations of broad a charter exempting it from selected state or local rules and
findings produced by SASS. regulations. A charter school may be a newly created
school or it may previously have been a public or private
Users of restricted-use SASS data can link school districts school.
and schools to other data sources. For instance, 2007–08
SASS restricted-use datasets include selected information BIE School. A school funded by the Bureau of Indian
taken from the CCD, but researchers can augment the Education of the Bureau of Indian Affairs, U.S.
datasets by adding more data from the CCD—either fiscal Department of the Interior. These schools may be operated
or nonfiscal data. by the BIE, a tribe, a private contractor, or an LEA.

3. KEY CONCEPTS Library Media Center. A library media center is an

organized collection of printed, audiovisual, or computer
Because of the large number of concepts in SASS surveys, resources that (a) is administered as a unit, (b) is located in
only those pertaining to the level of data collection (LEA, a designated place or places, and (c) makes resources and
school, teacher, library) are described in this section. For services available to students, teachers, and administrators.
additional terms, the reader is referred to glossaries in
SASS reports. Teacher. A full- or part-time teacher who teaches any
regularly scheduled classes in any of grades K–12. 2 This
Local Education Agency (LEA). A public school district, includes administrators, librarians, and other professional
or LEA, is defined as a government agency employing or support staff who teach regularly scheduled classes on a
elementary- and secondary-level teachers and part-time basis. Itinerant teachers are also included, as well
administratively responsible for providing public
elementary and/or secondary instruction and educational 2
A teacher teaching only kindergarten students is in scope,
support services. Districts that do not operate schools but
provided the school serves students in a grade higher than
employ teachers were last included in the 1999–2000 kindergarten.
SASS. (For example, some states have special education

SASS-4
 NCES Handbook of Survey Methods

as long-term substitutes who are filling the role of a Within each stratum, all non-BIE schools are
regular teacher on a long-term basis. An itinerant teacher is systematically selected using a probability proportionate to
one who teaches at more than one school (e.g., a music size algorithm. The measure of size used for schools in the
teacher who teaches 3 days per week at one school and 2 CCD is the square root of the number of teachers in the
days per week at another). Short-term substitute teachers school as reported in the CCD file. Any school with a
and student teachers are not included. measure of size larger than the sampling interval is
excluded from the probability sampling operation and
4. SURVEY DESIGN included in the sample with certainty.

Target Population The CCD Public Elementary/Secondary School Universe

LEAs that employ elementary- and/or secondary-level Survey serves as the public school sampling frame. (See
teachers (e.g., public school districts, state agencies that the CCD chapter for a more thorough discussion.) The
operate schools for special student populations, such as frame includes regular public schools, Department of
inmates of juvenile correctional facilities or students in Defense-operated military base schools, and special
Department of Defense schools); cooperative agencies that purpose schools (such as special education, vocational, and
provide special services to more than one school district; alternative schools). Schools outside the United States and
public, private, BIE, and charter schools with students in schools that teach only prekindergarten, kindergarten, or
any of grades 1–12; the principals of these schools; library postsecondary students are deleted from the file. The
media centers; and teachers in public, private, BIE, and following years of the CCD were used as the public school
charter schools who teach students in grades K–12 in a frame for the last five rounds of SASS:
school with at least a 1st grade.
• 2009–10 CCD for the 2011–12 SASS;
Sample Design
SASS uses a stratified probability sample design. Details • 2005–06 CCD for the 2007–08 SASS;
of stratification variables, sample selection, and frame • 2001–02 CCD for the 2003–04 SASS;
sources are provided below.
• 1997–98 CCD for the 1999–2000 SASS;
Public school sample. In the public school sample, schools
• 1991–92 CCD for the 1993–94 SASS; and
are selected first. The first level of stratification is by type
of school: (a) BIE schools (all BIE schools are • 1988–89 CCD for the 1990–91 SASS.
automatically in the sample); (b) schools with a high
percentage of American Indian students (i.e., schools with In the 1987–88 SASS, the 1986 Quality Education Data
19.5 percent or more American Indian students); (c) (QED) survey was used as the sampling frame.
schools in Delaware, Florida, Maryland, Nevada, and West
Private school sample. For private schools, the sample is
Virginia (where it is necessary to implement a different
stratified within each of the two types of frames: (1) a list
sampling methodology to select at least one school from
frame, which is the primary private school frame; and (2)
each LEA in the state); (d) charter schools; and (e) all
an area frame, which is used to identify schools not
other schools. Schools falling into more than one group are
included in the list frame and to compensate for the
assigned to types A, B, D, C, and E in that order. The
undercoverage of the list frame. Private schools in the list
second level of stratification varies within school type. All
frame are stratified by affiliation, grade level, and region.
BIE schools are automatically selected for the sample, so
Within each stratum, schools are sampled systematically
no stratification is needed. Schools with a high percentage
using a probability proportionate to size algorithm. Any
of American Indian students are stratified by state
school with a measure of size larger than the sampling
(Arizona; California; Montana; New Mexico; Washington;
interval is excluded from the probability sampling process
the remaining western states; Minnesota; North Dakota;
and included in the sample with certainty. All schools in
South Dakota; the remaining midwestern states; North
the area frame within noncertainty PSUs and not already
Carolina; Oklahoma; and the remaining states except
listed in the list frame are included in the sample with
Alaska, since most Alaskan schools have a high Native
certainty.
American enrollment). Schools in Delaware, Florida,
Maryland, Nevada, and West Virginia are stratified first by The most recent PSS, updated with the most recent
state and then by LEA. Charter schools and schools not association lists, serves as the private school sampling
placed in another category are stratified by state. Within frame. For example, the 2001–02 PSS—updated with 26
each second level, there are three grade level strata lists of private schools provided by a private school
(elementary, secondary, and combined schools). association (as well as 51 lists of private schools, from the
50 states and the District of Columbia)—was used as the
private school frame for the 2003–04 SASS. SASS 2011–

SASS-5
 NCES Handbook of Survey Methods

12 was based on the 2009–10 PSS. For the 2007–08 as early as possible in the 2007–08 school year at all public
SASS, the private school list frame was based on the (including public charter), private, and BIE-funded schools
2005–06 PSS, updated with private school organizations in the SASS sample to obtain a complete list of all the
and state lists collected by the U.S. Census Bureau in the teachers employed at each school.
summer of 2006. The 1991–92, 1989–90, and 1997–98
PSS were the basis for the private school frame for the In the 2007–08 SASS, teachers were stratified into one of
1993–94, 1990–91, and 1999–2000 SASS, respectively. two teacher types: new and experienced. For new and
The 1986 QED survey was used as the sampling frame for experienced teachers in public schools, oversampling was
the 1987–88 SASS. not required, due to the large number of sampled schools
with new teachers. Therefore, teachers were allocated to
BIE school selection. Since the 1993–94 SASS, all BIE the new and experienced categories in proportion to their
schools have been selected with certainty; in 1990–91, 80 numbers in the school. However, in private schools, new
percent of BIE schools were sampled. The BIE school teachers were oversampled. Before teachers were allocated
frame for the 2003–04 SASS consisted of a list of schools to the new or experienced strata, schools were first
that the BIE operated or funded during the 2001–02 school allocated an overall number of teachers to be selected.
year. (The list was obtained from the U.S. Department of
the Interior.) The BIE list was matched against the CCD, For the 2011–12 SASS, Teacher Listing Forms were
and the schools on the BIE list that did not match the CCD collected for sample schools and districts, and compiled by
were added to the universe of schools. the Census Bureau throughout the collection period.
Sampled schools provided information on teacher’s
For the 2007–08 SASS data collection, a separate universe teaching experience, with stratifications of beginning,
of schools operated or funded by the BIE in the 2005–06 early career, mid-year, and experienced. Begininning and
school year was drawn from the Program Education early career teachers were oversampled to improve survey
Directory maintained by the BIE. (The CCD now defines estimates for this subpopulation; within each teacher
the BIE as its own “territory,” similar to Puerto Rico and stratum within each school, teachers were selected
other non-state territories, and does not permit duplicates systematically with equal probability.
to be reported by the states.) All BIE schools meeting the
SASS definition of a school were included in the sample. Teacher records within a school are sorted by the teacher
stratum code, the teacher subject code, and the teacher line
After the 2007–08 SASS data collection however, BIE number code. The teacher line number code is a unique
data collection was discontinued; as a result no BIE number assigned to identify the teacher within the list of
schools, principals, teachers or library media centers were teachers keyed by the field representative. Within each
sampled for the 2011–12 SASS. teacher stratum in each school, teachers are selected
systematically with equal probability. The within-school
Charter school selection. In the 1999–2000 SASS, a probabilities of selection are computed so as to give all
charter school sample was added. All charter schools were teachers within a school stratum the same overall
selected with certainty from the frame, which consisted of probability of selection (self-weighted) within teacher and
a list of charter schools developed for the U.S. Department school strata, but not across strata. However, since the
of Education’s Institute of Education Sciences. The list school sample size of teachers is altered due to the
included only charter schools that were open (teaching minimum constraint (i.e., at least one teacher per school)
students) during the 1998–99 year. This changed in the or maximum constraint (i.e., no more than either twice the
2003–04 SASS, when a nationally representative sample average stratum allocation or 20 teachers per school), the
of public charter schools was included as part of the public goal of achieving self-weighting for teachers is lost in
school sample. In the 2011–12 SASS, charter schools some schools. Each sampled teacher receives a teacher
continued to be included as a part of the public school questionnaire.
sample.
Library media center selection. For the 2003–04 and
Each school sampled for SASS receives a school 2007–08 SASS, all library media centers in public, public
questionnaire, and the principal of each sampled school charter, and BIE-funded schools in the SASS sample were
receives a principal questionnaire. asked to complete the School Library Media Center
Questionnaire. For 2011–12, all library media centers in
Teacher selection. Within each sampled school, a sample public and public charter schools were sampled equating to
of teachers is selected. First, the sampled schools are asked roughly 10,250 public and 750 public charter library media
to provide a list of their teachers and selected centers being sampled.
characteristics. For example, in the 2007–08 SASS data
collection, the Teacher Listing Form was collected School district selection. In most states, once public
schools are selected, the districts associated with these

SASS-6
 NCES Handbook of Survey Methods

schools are placed in the sample as well. However, in schools to obtain completed questionnaires or to verify that
Delaware, Florida, Maryland, Nevada, and West Virginia, they have been mailed back. Field staff record the status of
all districts are defined as school sampling strata, placing each questionnaire and, if necessary, supply additional
all districts in each of these states in the district sample. (In blank questionnaires.
some SASS administrations, a sample of districts not
associated with schools is taken, but not in the 2007–08 Processing. During the check-in phase, each questionnaire
SASS.) For the 2011–12 SASS sampling frame, public is assigned an outcome code: completed interview, out-of-
charter schools were classified as dependent (governed by scope, or noninterview. A combination of manual data
a school district) and independent (not associated with a keying and imaging technology was used to enter the data.
school district). The district sample is selected using a Then, interview records in the data files undergo a round
probability proportionate to size algorithm. Each sampled of primary data review, where analysts examine the
school district receives a school district questionnaire. The frequencies of each data item in order to identify any
approximate sample sizes for the 2011–12 SASS were suspicious values. Census staff review the problem cases
51,100 public school teachers; 7,100 private school and make corrections whenever possible.
teachers; 14,000 school principals; and 5,800 school
districts. After the primary data review, all records (i.e., records
from all survey components) classified as interviews are
Data Collection and Processing subject to a set of computer edits: a range check, a
In the 2011–12 SASS, teachers were mailed an invitation consistency edit, and a blanking edit. After the completion
to complete a web-based questionnaire, although they of these edits, the records are put through another edit to
could also request to take the paper questionnaire. 67% of make a final determination of whether the case is eligible
public school teachers and 59% of private school teachers for the survey, and, if so, whether sufficient data have been
chose to use the web-based questionnaire. In 2003–04 and collected for the case to be classified as an interview. A
2007–08, the School Library Media Center Survey did not final interview status recode (ISR) value is assigned to
have an Internet reporting option, as it did in 1999–2000. each case as a result of the edit.
All survey modes used in SASS are administered by the
U.S. Bureau of the Census. Estimation Methods
Sample units are weighted to produce national and state
Reference Dates. Data for SASS components are collected estimates for public elementary and secondary school
during a single school year. Most data items refer to that surveys (i.e., schools, teachers, administrators, school
school year. Questions on enrollment and staffing refer to districts, and school library media centers); and national
October 1 of the school year. Questions for teachers about estimates for BIE, charter school, and public combined
current teaching loads refer to the most recent full week school surveys (i.e., schools, teachers, administrators, and
that school was in session, and questions on professional school library media centers). The private sector is
development refer to the past 12 months. weighted to produce national and affiliation group
estimates. These estimates are produced through the
Data Collection. The data collection procedures begin weighting and imputation procedures discussed below.
with advance mailings to school districts explaining the
nature and purpose of SASS. Field staff attempt to Weighting. Estimates from SASS sample data are
establish a contact person for the School District produced by using weights. The weighting process for
Questionnaire and determine whether the district is willing each component of SASS includes adjustments for
and able to provide an electronic list of teachers for their nonresponse using respondents’ data and adjustments of
selected school(s) in the fall. If the district agrees to the sample totals to the frame totals to reduce sampling
provide an electronic list, field staff determine the variability. The exact formula representing the
appropriate contact person to receive the request. Field construction of the weight for each component of SASS is
staff verifiy the selected schools’ names, grade ranges, and provided in each administration’s sample design report
operational statuses. Finally, field staff attempt to collect (e.g., 1993–94 Schools and Staffing Survey: Sample
the names of the selected schools’ principals and their e- Design and Estimation [Abramson et al. 1996]). The
mail addresses. construction of weights is also discussed in the Quality
Profile reports (Jabine 1994; Kalton et al. 2000) and in the
The school district questionnaires are mailed out first. documentation for the 2003–04 administration (Tourkin et
Then, the school, principal, and library media center al. 2007). Since SASS and PSS data were collected at the
surveys are delivered to schools in person. The teacher same time in 1993–94 and 1999–2000, in both years the
questionnaires are delivered last. Follow-up efforts begin number of private schools reported in SASS was made to
approximately 2 weeks after questionnaires are distributed. match the number of private schools reported in PSS.
They consist of telephone calls and personal visits to

SASS-7
 NCES Handbook of Survey Methods

Imputation. In all administrations of SASS, all items with “other.” Nonsectarian schools were divided into three
missing values are imputed for records classified as groups by typology.
interviews. SASS uses a two-stage imputation procedure.
The first-stage imputation uses a logical or deductive • Grade-level stratification in public and private schools
method, such as: was defined purely on the basis of grade level of the
school starting in 2003–04 SASS. Schools classified as a
• Using data from other items in the same questionnaire; type other than “regular school” were no longer placed
by default in the combined school category, which
• Extracting data from a related SASS component includes schools with some elementary and some
(different questionnaire); or secondary grades. Many nonregular schools (i.e., special
education, alternative, and vocational schools) cover a
• Extracting information about the sample case from the
specific grade range. To the extent this grade range is
PSS or CCD, the sampling frames for private and public
known, this was a more appropriate method of
schools, respectively.
stratification than placing them all in the combined
In addition, some inconsistencies between items are school strata. Nonregular schools with a grade range that
corrected by ratio adjustment during the first-stage is ungraded or unknown remain in the combined school
imputation. strata.

The second-stage imputation process is applied to all items • Public schools from the CCD were collapsed into what
with missing values that were not imputed in the first was perceived to be a better fit with the SASS definition
stage. This imputation uses a hot-deck imputation method, of a school prior to stratification beginning in the 2003–
extracting data from a respondent (i.e., a donor) with 04 SASS. The sample allocation was revised to avoid
similar characteristics to the nonrespondent. If there is still undersampling schools now classified at the combined
no observed value after collapsing to a certain point, the grade level. In other words, the revision of the sample
missing values are imputed using a clerically imputed allocation ensured that the newly combined schools
value or automated algorithm. were sampled at the same approximate rate as they
would have been prior to the collapsing procedure. In
Recent Changes general, the combined school sample size was increased
Several changes were made over time, largely due to to the point at which the combined school sampling rate
budgetary reasons. equaled the overall state-level sampling rate. For
example, if one in five schools were sampled in a
Design changes from 1999–2000 to 2011–12: particular state, then one in five of the combined schools
were sampled rather than using the default sample size
• Since the 2007–08 SASS, BIE schools, principals, of 10 combined schools.
teachers, or library media centers are no longer sampled.
• The sort order for the public and private school sampling
• Rather than surveying all public charter schools, as was was altered to sort on enrollment in a serpentine fashion
done in the 1999–2000 SASS, some 300 public charter (instead of always sorting in descending order) in the
schools were sampled for the 2003–04 SASS. 2003–04 SASS. Serpentine sorting involves sorting in
• The separate questionnaire for public charter schools ascending order with respect to higher level sort
was discontinued. The reduction in the public charter variables one time, then sorting in descending order the
school sample size from 1,100 in the 1999–2000 SASS next time, and so on. This reduces the variation in
to about 300 in the 2003–04 SASS meant it was no enrollment between adjacent sampled schools and thus
longer feasible to produce a separate questionnaire, since reduces the overall sampling error.
public charter school data could not be published with as • Florida and Maryland were added to the list of states
much detail (for the 2003–04 SASS, only at the national where at least one school is selected in each school
and regional levels). Public charter school data are now district. This was done in the 2003–04 SASS to decrease
included with traditional public school data. the standard error of the state-level school district
• Affiliation for private schools was redefined and estimates.
stratified into 17 groups rather than the previous 20 • Oversampling of bilingual/English as a Second
groups in the 2003–04 SASS. Catholic schools were Language (ESL) teachers was discontinued in the 2003–
split into three groups based on typology. Other 04 SASS, since a sufficient number of bilingual teachers
religious schools were divided into five groups to produce the desired reliability estimates could be
corresponding to the four largest non-Catholic religious done without oversampling.
organizations (by number of schools) and a catch-all

SASS-8
 NCES Handbook of Survey Methods

• Teacher sampling was automated to speed up the bootstrap frame, replicate weights computed, and
distribution of the teacher questionnaires. This, however, variances estimated with standard BHR software. The
reduced the level of control over the sample sizes for the bootstrap replicate basic weights (inverse of the probability
remaining oversampled teacher strata (Asian/Pacific of selection) were subsequently reweighted. More
Islander and American Indian/Alaska Native). The information on the bootstrap variance methodology and
automation no longer allowed the sampling rate for how it applies to SASS is contained in the following
these teachers to be periodically revised during the sources: “A Bootstrap Variance Estimator for Systematic
sampling process. Thus, if the number of these teachers PPS Sampling” (U.S. Department of Education 2000)
listed differed from the expected number, the sample which describes the methodology used in the 1999–2000
size goal would no longer be met. SASS; “A Bootstrap Variance Estimator for the Schools
and Staffing Survey” (U.S. Department of Education
• The School Library Media Center Questionnaire was not 1994); “Balanced Half-Sample Replication With
administered to private schools for budget reasons as of Aggregation Units” (U.S. Department of Education 1994);
the 2003–04 SASS. “Comparing Three Bootstrap Methods for Survey Data”
• The School Questionnaire (with district items) is a (Sitter 1990); “Properties of the Schools and Staffing
questionnaire that contains the public school questions Survey Bootstrap Variance Estimator” (U.S. Department
and most of the school district questions in the 1999– of Education 1996); and “The Jackknife, the Bootstrap and
2000 SASS. It was administered to public charter, state- Other Resampling Plans” (Efron 1982).
operated (often schools for the blind or schools located
in juvenile detention facilities), and BIE-funded schools, SASS variances can be calculated using the replicates of
as well as public schools in one-school districts. This the full sample that are available in the data files with
change was made to ease respondent burden in cases software such as WesVarPC. For examples of other
where the respondent for the school and school district software that support BRR, see Introduction to Variance
questionnaires was expected to be the same. Estimation (Wolter 1985).

Future Plans Average Design Effects. Design effects (Deffs) measure

SASS was re-designed into the National Teacher and the impact of the complex sample design on the accuracy
Principal Survey (NTPS), which was administered for the of a sample estimate, in comparison to the alternative
first time in 2015–16. simple random sample design. For the 1990–91 SASS, an
average design effect was derived for groups of statistics
5. DATA QUALITY AND and, within each group, for a set of subpopulations.
COMPARABILITY Standard errors for 1990–91 and 1993–94 SASS statistics
of various groups for various subpopulations can then be
Sampling Error calculated approximately from the standard errors based on
The estimators of sampling variances for SASS statistics the simple random sample (using SAS or SPSS) in
take the SASS complex sample design into account. For an conjunction with the average design effects provided. For
overview of the calculation of sampling errors, see the example, for the 1990–91 SASS, average design effects for
Quality Profile reports (Jabine 1994; Kalton et al. 2000). selected variables in the School Survey are 1.60 (public
sector) and 1.36 (private sector); in the Principal Survey,
Direct Variance Estimators. The balanced half-sample 4.40 (public sector) and 4.02 (private sector); and in the
replication (BHR) method, also called balanced repeated Teacher Survey, 3.75 (public sector) and 2.52 (private
replication (BRR), was used to estimate the sampling sector). Examples illustrating the use of SASS average
errors associated with estimates from the 1987–88 and design effect tables are provided in Design Effects and
1990–91 SASS. Given the replicate weights, the statistic of Generalized Variance Functions for the 1990–91 Schools
interest (e.g., the number of 12th grade teachers from the and Staffing Survey (SASS), Volume I, User’s Manual
School Survey) can be estimated from the full sample and (Salvucci and Weng 1995).
from each replicate. The mean square error of the replicate
estimates around the full sample estimate provides an Generalized Variance Functions (GVFs). GVF tables
estimate of the variance of the statistic. were developed for use in the calculation of standard errors
of totals, averages, and proportions of interest in the 1990–
A bootstrap variance estimator was used for the 1993–94, 91 SASS components. The 1990–91 GVFs can be used for
1999–2000, 2003–04, 2007–08, and 2011–12 SASS. The the 1993–94 SASS because no major design changes were
bootstrap variance reflects the increase in precision due to adopted between 1990–91 and 1993–94. Note that the
large sampling rates because the bootstrap is done GVF approach, unlike the design effect approach
systematically without replacement, as was the original described above, involves no need to calculate the simple
sampling. Bootstrap samples can be selected from the random sample variance estimates. Examples illustrating

SASS-9
 NCES Handbook of Survey Methods

the use of the GVF tables are provided in Design Effects of Response Rates in the 1990–91 Schools and Staffing
and Generalized Variance Functions for the 1990–91 Survey (SASS) (Scheuren et al. 1996).
Schools and Staffing Survey (SASS), Volume I, User’s
Manual (Salvucci and Weng 1995). Item Nonresponse. For the 2011–12 SASS, the percentage
with weighted item response rates at or above 85 percent,
Nonsampling Error by individual survey, are: 100 percent for public school
Coverage Error. SASS surveys are subject to any districts; 96 percent for public schools; 94 percent for
coverage error present in the CCD and PSS data files, private schools; 99 percent for public school principals; 98
which serve as their principal sampling frames. The report percent for private school principals; 94 percent for public
Coverage Evaluation of the 1994–95 Common Core of school teachers; and 93 percent for private school teachers.
Data: Public Elementary/Secondary Education Agency Only private school teachers surveys had items with
Universe Survey (Owens 1997) found that overall coverage response rates below 70 percent.
in the 1994–95 CCD Local Education Agency Universe
Survey was 96.2 percent (in a comparison to state Measurement Error. Results reported in An Analysis of
education directories). “Regular” agencies—those Total Response in the 1993–94 Schools and Staffing
traditionally responsible for providing public education— Survey (SASS) (Monaco et al. 1997) support the contention
had almost total coverage in the 1994–95 agency universe that, without follow-up to mail surveys, nonresponse error
survey. Most coverage discrepancies were attributed to would be much greater than it is and that the validity and
nontraditional agencies that provide special education, reliability of the data would be considerably reduced.
vocational education, and other services. However, there is However, because of the substantial amount of telephone
potential for undercoverage bias associated with the follow-up, there is concern about possible bias due to
absence of schools built between the time when the differences in the mode of survey collection. Other
sampling frame is constructed and the time of the SASS possible sources of measurement error include long,
survey administration. Further research on coverage can be complex instructions that respondents either do not read or
found in Evaluating the Coverage of the U.S. National do not understand, navigation problems related to the
Center for Education Statistics’ Public format of the questionnaires, and definitional and
Elementary/Secondary School Frame (Hamann 2000) and classification problems. See also Measurement Error
Evaluating the Coverage of the U.S. National Center for Studies at the National Center for Education Statistics
Education Statistics’ Public and Private School Frames (Salvucci et al.1997).
Using Data from the National Assessment of Educational
Progress (Lee, Burke, and Rust 2000). Several NCES working papers also address measurement
error. Reports on the 1993–94 SASS include Cognitive
A capture-recapture methodology was used to estimate the Research on the Teacher Listing Form for the Schools and
number of private schools in the United States and to Staffing Survey (Jenkins and Von Thurn 1996); Further
estimate the coverage of private schools in the 1999–2000 Cognitive Research on the Schools and Staffing Survey
PSS; the study found that the PSS school coverage rate is (SASS) (Zukerberg and Lee 1997); Report of Cognitive
equal to 97 percent. (See CCD and PSS chapters for a Research on the Public and Private School Teacher
more thorough discussion.) Questionnaires for the Schools and Staffing Survey 1993–
94 School Year (Jenkins 1997); and Response Variance in
Nonresponse Error. the 1993–94 Schools and Staffing Survey: A Reinterview
Unit nonresponse. The weighted unit response rates for Report (Bushery, Schreiner, and Sebron 1998). Reports on
public schools have been higher than the weighted unit the 1991–92 SASS include the 1991 Schools and Staffing
response rates for private schools in all six rounds of Survey (SASS) Reinterview Response Variance Report
SASS. (See table SASS-1 for response rates from selected (Royce 1994) and The Results of the 1991–92 Teache
years.) For more information on the analysis of Follow-up Survey (TFS) Reinterview and Extensive
nonresponse rates, refer to An Analysis of Total Reconciliation (Jenkins and Wetzel 1995).
Nonesponse in the 1993–94 Schools and Staffing Survey
(SASS) (Monaco et al. 1997) and An Exploratory Analysis

SASS-10
 NCES Handbook of Survey Methods

Table SASS-1. Summary of weighted unit response rates for selected SASS questionnaires: 1993–94 through 2011–12

Questionnaire 1993–94 1999–2000 2003–04 2007–08 2011–12

School District Survey 93.9 88.6 82.9 87.8 80.6
Public Principal Survey 96.6 90.0 82.2 79.4 72.7
Public School Survey 92.3 88.5 80.8 80.4 72.5
Public Teacher Survey1 88.2 92.2 89.2 86.2 79.6
Private Principal Survey 87.6 84.8 74.9 72.2 64.7
Private School Survey 83.2 79.8 75.9 75.9 65.7
Private Teacher Survey1 80.2 87.0 85.4 85.1 71.6
BIE Principal Survey 98.7 93.3 90.7 79.2 —
BIE School Survey 99.3 96.7 89.5 77.1 —
BIE Teacher Survey 86.5 97.8 93.8 87.3 —
—Not available. Data were not collected.
1
The overall teacher response rates are the percentage of teachers responding in schools that provided teacher lists for
sampling.
SOURCE: SASS methodology reports; available at https://nces.ed.gov/pubsearch/getpubcats.asp?sid=003.

6. CONTACT INFORMATION Education. Washington, DC.

http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=911
For content information on SASS, contact: 37.

Maura Spiegelman Broene, P. (1991b.) User’s Manual: 1987–88 Schools and

Phone: (202) 245-6581 Staffing Survey - Public School Data (NCES 91-136).
E-mail: [email protected] National Center for Education Statistics, U.S.
Department of Education. Washington, DC.
Mailing Address http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=911
National Center for Education Statistics 36.
Institute of Education Sciences
Potomac Center Plaza Cox, S., Parmer, R., Strizek, G., and Thomas, T. (2016).
550 12th Street, SW Documentation for the 2011–12 Schools and Staffing
Washington, DC 20202 Survey (NCES 2016-817). National Center for
Education Statistics, U.S. Department of Education.
7. METHODOLOGY AND EVALUATION Washington, DC.
REPORTS https://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=20
16817.
General
Aritomi, P., and Coopersmith, J. (2009). Characteristics of Goldring, R., Taie, S. Rizzo, L. Colby, D. and Fraser, A.
Public School Districts in the United States: Results (2013). User’s Manual for the 2011-12 Schools and
From the 2007-08 Schools and Staffing Survey (NCES Staffing Survey Volume 1: Overview (NCES 2013-
2009-320). National Center for Education Statistics, 330). National Center for Education Statistics, U.S.
Institute of Education Sciences, U.S. Department of Department of Education. Washington, DC.
Education. Washington, DC. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=201
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2009 4356.
320.
Gruber, K.J., Rohr, C.L., and Fondelier, S.E. (1994).
Broene, P. (1991a). User’s Manual: 1987–88 Schools and 1990–91 Schools and Staffing Survey: Data File
Staffing Survey - Public School Administrator User’s Manual Volumes I-IV: Public Use Codebook
Questionnaire Data (NCES 91-137). National Center and Bureau of Indian Affairs Codebooks (NCES 93-
for Education Statistics, U.S. Department of 144). National Center for Education Statistics, U.S.

SASS-11
 NCES Handbook of Survey Methods

Department of Education. Washington, DC. Mullens, J.E., and Kasprzyk, D. (1997). The Schools and
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=931 Staffing Survey: Recommendations for the Future
44. (NCES 97-596). National Center for Education
Statistics, U.S. Department of Education. Washington,
Gruber, K.J., Rohr, C.L., and Fondelier, S.E. (1996). DC.
1993–94 Schools and Staffing Survey: Data File http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=975
User’s Manual, Volume I: Survey Documentation 96.
(NCES 96-142). National Center for Education
Statistics, U.S. Department of Education. Washington, Wiley, S.D., and Reynolds, K.A. (1999). Tracking
DC. Secondary Use of the Schools and Staffing Survey
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=961 Data: Preliminary Results (NCES Working Paper
42. 1999-02). National Center for Education Statistics,
U.S. Department of Education. Washington, DC.
Rouk, Ü., Weiner, L., and Riley, D. (1999). What Users http://nces.ed.gov/pubs99/199902.pdf.
Say About Schools and Staffing Survey Publications
(NCES Working Paper 1999-10). National Center for Survey Design
Education Statistics, U.S. Department of Education. Abramson, R., Cole, C., Fondelier, S., Jackson, B., Parmer,
Washington, DC. R., and Kaufman, S. (1996). 1993–94 Schools and
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=199 Staffing Survey: Sample Design and Estimation
910. (NCES 96-089). National Center for Education
Statistics, U.S. Department of Education. Washington,
Tourkin, S.C., Pugh, K.W., Fondelier, S.E., Parmer, R.J., DC.
Cole, C., Jackson, B., Warner, T., and Weant, G. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=960
(2004). 1999–2000 Schools and Staffing Survey 89.
(SASS) Data File User’s Manual (NCES 2004-303).
National Center for Education Statistics, Institute of Baker, D.P., Levine, R., Han, M., and Garet, M. (1998). A
Education Sciences, U.S. Department of Education. Feasibility Study of Longitudinal Design for Schools
Washington, DC. and Staffing Survey (NCES Working Paper 98-16).
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=200 National Center for Education Statistics, U.S.
4303. Department of Education. Washington, DC.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=981
Tourkin, S., Thomas, T., Swaim, N., Cox, S., Parmer, R., 6.
Jackson, B., Cole, C., and Zhang, B. (2010).
Documentation for the 2007-08 Schools and Staffing Ingersoll, R.M. (1995). An Agenda for Research on
Survey (NCES 2010-332). National Center for Teachers and Schools: Revisiting NCES’ Schools and
Education Statistics, Institute of Education Sciences, Staffing Survey (NCES Working Paper 95-18).
U.S. Department of Education. Washington, DC. National Center for Education Statistics, U.S.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=201 Department of Education. Washington, DC.
0332. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=951
8.
Tourkin, S.C., Warner, T., Parmer, R., Cole, C., Jackson,
B., Zukerburg, A., Cox, S., and Soderborg, A. (2007). Ingersoll, R.M. (1996). National Assessments of Teacher
Documentation for the 2003–04 Schools and Staffing Quality (NCES Working Paper 96-24). National
Survey (NCES 2007-337). National Center for Center for Education Statistics, U.S. Department of
Education Statistics, Institute of Education Sciences, Education. Washington, DC.
U.S. Department of Education. Washington, DC. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=962
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=200 4.
7337.
Isaacs, J.B., Best, C.M., Cullen, A.D., Garet, M.S., and
Uses of Data Sherman, J.D. (1998). Collection of Public School
McGrath, D.J., and Luekens, M.T. (2000). A Research Expenditure Data: Development of a Questionnaire
Agenda for the 1999–2000 Schools and Staffing (NCES Working Paper 98-01). National Center for
Survey (NCES Working Paper 2000-10). National Education Statistics, U.S. Department of Education.
Center for Education Statistics, U.S. Department of Washington, DC.
Education. Washington, DC. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=980
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=200 1.
010.

SASS-12
 NCES Handbook of Survey Methods

Isaacs, J.B., Garet, M.S., Sherman, J.D., Cullen, A., and Elementary/Secondary School Frame. In The Second
Phelps, R. (1999). Collection of Resource and International Conference on Establishment Surveys:
Expenditure Data on the Schools and Staffing Survey Survey Methods for Businesses, Farms, and
(NCES Working Paper 1999-07). National Center for Institutions (pp. 79–88). Arlington, VA: American
Education Statistics, U.S. Department of Education. Statistical Association.
Washington, DC.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=199 Jabine, T. (1994). Quality Profile for SASS: Aspects of the
907. Quality of Data in the Schools and Staffing Surveys
(SASS) (NCES 94-340). National Center for Education
Kaufman, S. (1991). 1988 Schools and Staffing Survey Statistics, U.S. Department of Education. Washington,
Sample Design and Estimation (NCES 91-127). DC.
National Center for Education Statistics, U.S. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=943
Department of Education. Washington, DC. 40.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=911
27. Jenkins, C.R. (1997). Report of Cognitive Research on the
Public and Private School Teacher Questionnaires for
Kaufman, S., and Huang, H. (1993). 1990–91 Schools and the Schools and Staffing Survey 1993–94 School Year
Staffing Survey: Sample Design and Estimation (NCES Working Paper 97-10). National Center for
(NCES 93-449). National Center for Education Education Statistics, U.S. Department of Education.
Statistics, U.S. Department of Education. Washington, Washington, DC.
DC. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=971
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=934 0.
49.
Jenkins, C.R., and Von Thurn, D. (1996). Cognitive
Levine, R.E., Chambers, J.G., Dueñas, I.E., and Hikido, Research on the Teacher Listing Form for the Schools
C.S. (1997). Improving the Measurement of Staffing and Staffing Survey (NCES Working Paper 96-05).
Resources at the School Level: The Development of National Center for Education Statistics, U.S.
Recommendations for NCES for the Schools and Department of Education. Washington, DC.
Staffing Survey (NCES Working Paper 97-42). http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=960
National Center for Education Statistics, U.S. 5.
Department of Education. Washington, DC.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=974 Jenkins, C.R., and Wetzel, A. (1995). The Results of the
2. 1991–92 Teacher Follow-up Survey (TFS)
Reinterview and Extensive Reconciliation (NCES
Rollefson, M. (1998). The Redesign of the Schools and Working Paper 95-10). National Center for Education
Staffing Survey for 1999–2000: A Position Paper Statistics, U.S. Department of Education. Washington,
(NCES Working Paper 98-08). National Center for DC.
Education Statistics, U.S. Department of Education. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=951
Washington, DC. 0.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=980
8. Kalton, G., Winglee, M., Krawchuk, S., and Levine, D.
(2000). Quality Profile for SASS Rounds 1–3: 1987–
Data Quality and Comparability 1995, Aspects of the Quality of Data in the Schools
Bushery, J., Schreiner, I., and Sebron, G. (1998). Response and Staffing Surveys (SASS) (NCES 2000-308).
Variance in the 1993–94 Schools and Staffing Survey: National Center for Education Statistics, U.S.
A Reinterview Report (NCES Working Paper 98-02). Department of Education. Washington, DC.
National Center for Education Statistics, U.S. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=200
Department of Education. Washington, DC. 0308.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=980
2. Lee, H., Burke, J., and Rust, K. (2000). Evaluating the
Coverage of the U.S. National Center for Education
Efron, B. (1982). The Jackknife, the Bootstrap, and Other Statistics’ Public and Private School Frames Using
Resampling Plans. Society of Industrial and Applied Data From the National Assessment of Educational
Mathematics CBMS-NSF Monographs, 38. Progress. In The Second International Conference on
Establishment Surveys: Survey Methods for
Hamann, T.A. (2000). Evaluating the Coverage of the U.S. Businesses, Farms, and Institutions (pp. 89–98).
National Center for Education Statistics’ Public Arlington, VA: American Statistical Association.

SASS-13
 NCES Handbook of Survey Methods

Monaco, D., Salvucci, S., Zhang, F., and Hu, M. (1997). Sitter, R. (1990). Comparing Three Bootstrap Methods for
An Analysis of Total Nonesponse in the 1993–94 Survey Data. Technical Report Series of the
Schools and Staffing Survey (SASS) (NCES 98-243). Laboratory for Research in Statistics and Probability.
National Center for Education Statistics, U.S. Carleton University.
Department of Education. Washington, DC.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=982 U.S. Department of Education, National Center for
43. Education Statistics. (1994). A Bootstrap Variance
Estimator for the Schools and Staffing Survey;
Owens, S. (1997). Coverage Evaluation of the 1994–95 Balanced Half-Sample Replication With Aggregation
Common Core of Data: Public Elementary/Secondary Units. In Schools and Staffing Survey (SASS), Papers
Education Agency Universe Survey (NCES 97-505). Presented at the Meetings of the American Statistical
National Center for Education Statistics, U.S. Association (NCES Working Paper 94-01).
Department of Education. Washington, DC. Washington, DC: Author.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=975 http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid
05. =9401.

Royce, D. (1994). 1991 Schools and Staffing Survey U.S. Department of Education, National Center for
(SASS) Reinterview Response Variance Report (NCES Education Statistics. (1996). Properties of the Schools
94-03). National Center for Education Statistics, U.S. and Staffing Survey Bootstrap Variance Estimator. In
Department of Education. Washington, DC. Schools and Staffing Survey (SASS): 1995 Selected
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=940 Papers Presented at the 1995 Meeting of the
3. American Statistical Association (NCES Working
Paper 96-02). Washington, DC: Author.
Salvucci, S., Conley, V., Fink, S., and Saba, M. (1997). http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=960
Measurement Error Studies at the National Center for 2.
Education Statistics (NCES 97-464). National Center
for Education Statistics, Sciences, U.S. Department of U.S. Department of Education, National Center for
Education. Washington, DC. Education Statistics. (2000). A Bootstrap Variance
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=974 Estimator for Systematic PPS Sampling. In Selected
64. Papers on Education Surveys: Papers Presented at
the 1998 and 1999 ASA and 1999 AAPOR Meetings
Salvucci, S., Holt, A., and Moonesingle, R. (1995). Design (NCES Working Paper 2000-04). Washington, DC:
Effects and Generalized Variance Functions for the Author.
1990–91 Schools and Staffing Survey (SASS): Volume http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=200
II, Technical Report (NCES 95-342-II). National 004.
Center for Education Statistics, U.S. Department of
Education. Washington, DC. Wolter, K.M. (1985). Introduction to Variance Estimation.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=953 New York: Springer-Verlag.
42-II.
Zukerberg, A., and Lee, M. (1997). Further Cognitive
Salvucci, S., and Weng, S. (1995). Design Effects and Research on the Schools and Staffing Survey (SASS)
Generalized Variance Functions for the 1990–91 Teacher Listing Form (NCES Working Paper 97-23).
Schools and Staffing Surveys (SASS) Volume I— National Center for Education Statistics, U.S.
User’s Manual (NCES 95-342-I). National Center for Department of Education. Washington, DC.
Education Statistics, U.S. Department of Education. http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=972
Washington, DC. 3.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=953
42-I.

Scheuren, F., Monaco, D., Zhang, F., Ikosi, G., and Chang,
M. (1996). An Exploratory Analysis of Response Rates
in the 1990–91 Schools and Staffing Survey (SASS)
(NCES 96-338). National Center for Education
Statistics, U.S. Department of Education. Washington,
DC.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=963
38.

SASS-14