ERIC Identifier: ED383518
Publication Date: 1995-05-00
Author: Huang, Gary
Source: ERIC Clearinghouse on Rural
Education and Small Schools Charleston WV.
National Data for Studying Rural Education: Elementary and
Secondary Education Applications. ERIC Digest.
Information collected specifically on rural education is scant (Haas, 1992;
Stern, 1994). However, the National Center for Education Statistics (NCES)
recently geared up its efforts to disseminate data and products on elementary
and secondary education that contain measures of community urbanicity, making
rural-urban comparisons or rural-focused analyses possible (Stephens, 1992).
This digest describes NCES datasets, presents issues that can be addressed with
NCES data, and offers practical tips for accessing these data.
NCES has primarily two types of data: population data and survey data.
Population data cover the nation's school universe and provide descriptive
information. Survey data do not actually cover the national population, but can
yield estimates from nationally representative samples of schools or students.
Most NCES surveys are conducted by questionnaire (some with supplementary
telephone or personal interviews) using stratified probability samples. Driven
by specific policy issues, surveys collect detailed and dynamic information.
The Common Core of Data (CCD) covers all
public elementary and secondary schools as well as local and state education
agencies. Information collected by CCD includes descriptive data on schools and
districts (name, address, phone number, and locale); demographic data on
students and staff; and fiscal data (revenues and current expenditures). CCD is
a major source for identifying and describing public elementary and secondary
schools and school districts in the U.S.
The School District Data Book (SDDB) provides the most comprehensive data on
the universe of school districts and communities (Herriot, 1992). Installed in a
set of CD-ROMs, SDDB incorporates the Census Bureau's 1990 decennial data with
data from NCES's CCD and School and Staffing Survey (described below). SDDB can
(1) provide education and demographic profiles of the nation, states, counties,
and school districts; (2) tabulate a variety of statistics at the levels of the
census geographies, household, and the child; and (3) display thematic maps of
educational and demographic conditions of the nation, states, counties, and
districts, allowing users to display data they themselves have manipulated. SDDB
will be useful for both research and program planning in rural education since
it links sociodemographic complexities to schooling.
SURVEY DATA--LONGITUDINAL STUDIES
NCES longitudinal surveys
follow the same sample of respondents across a period of years to see how
education processes evolve. The three projects described below are similar in
sampling design, data collection procedures, and measurement. They all include
multiple files with data gathered from students, teachers, schools, and parents,
but reflect changes in substantive concerns. Together, they provide
opportunities for understanding the life course of American youth in three
decades, taking into account a host of individual, family, and school processes
(NCES, in press).
The National Longitudinal Survey (NLS72) contains information on a nationally
representative sample of 12th graders enrolled in the 1971-72 school year, with
four follow-ups to 1986. The main focus is the school-to-work transition.
The High School and Beyond Survey (HS&B) began in 1980 and has followed a
senior cohort through 1986 and a sophomore cohort through 1992. Data from the
study supports the study of high school experiences and the subsequent life
course followed by respondents. It addressed issues arising in the 1980s such as
declining test scores, climbing dropout rates, the goals of vocational
education, and access to postsecondary education.
The National Education Longitudinal Study of 1988 (NELS:88) is an ongoing
project that builds upon existing knowledge about the impact of middle school
and high school educational achievement by following students beginning in
eighth grade. Starting with the 1988 base year survey of eighth grade students,
NELS:88 completed the third follow-up survey in 1994 using refined sampling
strategies, and will continue to follow this cohort through 1998.
SURVEY DATA--CROSS-SECTIONAL STUDIES
surveys study different samples or cohorts to examine the conditions at a given
The School and Staffing Survey (SASS) was conducted during the 1987-88 and
1990-91 school years, and will be conducted every three years in the future.
Collecting information on schools, districts, and administrators, SASS is the
most comprehensive database for studying the work force of teachers and
administrators in both public and private schools.
The National Assessment of Educational Progress (NAEP) is a congressionally
mandated project that has collected and reported information for over 25 years
on what American students know and can do. NAEP provides objective data on
student performance at national and regional levels in reading, mathematics,
science, writing, citizenship, U.S. history, geography, social studies, art,
music, literature, computer competence, and career and occupational development.
The National Household Education Survey (NHES) is NCES's only household
survey. The 1991 NHES focuses on the utilization and the condition of child care
services and adult education participation. Data were collected by conducting
telephone interviews of a national sample of households. The 1993 project looks
at the issues of school readiness and school safety/discipline. Surveys are also
planned for 1995 and 1996.
USING NCES DATA IN PROGRAM PLANNING
In designing a program,
planners need to look at local strengths and weaknesses compared to other
communities. NCES's school population data can fulfill such needs. SDDB allows
comparisons between a given district to other districts in the state or in the
nation in specified aspects of demography and education. It enables elaborate
but easy-to-operate analysis because of the comprehensiveness of the data and
the flexibility in data manipulation (e.g., breaking down data by
characteristics of children, parents, households, and school districts to
produce profiles, tables, and maps). The results may help formulate plans to
meet local needs and convince stakeholders to support particular projects. A
concern in regard to such local application is the currency of the decennial
census data. Since the 1990 census will not be updated until 2000, the reality
in areas with highly mobile demography may soon differ from that portrayed by
CCD can serve some of these functions, though the data elements are more
narrowly focused. Annually updated, CCD is useful for descriptive and analytic
studies of schools and school districts. It is also an authority source for
sampling at the national and local levels for marketing, polling, and survey
Besides population data, some surveys (SASS, NAEP, NHES) also can serve for
local planning because (1) they produce reasonably good estimates of regional or
local conditions and (2) they are conducted periodically to generate timely
USING NCES DATA IN POLICY-MAKING
knowledge about program effectiveness and local needs for resources, incentives,
and technical assistance. NCES data can address, under the rubric of school
effectiveness, issues such as rural-urban differences in operation and
performance; rural school financial conditions; course offering and taking; and
working conditions of rural administrators and teachers and their consequences
to student outcomes.
While population data such as SDDB can tell some on these issues, more
dynamic information is available from survey projects such as SASS, whose
sampling design allows rural-urban comparisons in particular states, regions,
and the nation as a whole. NAEP is another source for investigating the
effectiveness issue in urban and rural education. The power of NAEP is its
continuity and extensive coverage of performance data, which yield great
potential for trend analysis. But its complexities in sampling and measurement
may be a challenge.
Inadequate coverage of rural-specific policy issues by NCES datasets is a
difficulty facing rural policy research. School consolidation, an issue with
high stakes to policymaking and local community life, has evoked ongoing debate
in many rural communities (Hobbs, 1991). Yet, NCES datasets do not contain
information on the history or processes of school consolidation. Another
unaddressed rural issue is local components in curriculum. Community-based
school programs such as Foxfire are said to benefit students and rural
communities by cultivating meanings of rural life (Wigginton, 1985; Theobald,
1992). Increasingly, rural schools are incorporating the approach into their
programs (Hobbs, 1991; Stern, 1994). However, systematic studies of such
programs' effectiveness are impossible with NCES datasets because they have not
yet gathered information about innovative rural-specific curricula.
USING NCES DATA IN SCHOLARLY RESEARCH
tries to understand education processes in a more general sense, resulting in
findings that may or may not have direct implications for current policy issues.
For instance, a dilemma in rural education interesting to researchers is that
while rural communities badly need an educated population and have limited
resources for education, schools continue to train students in urban-oriented
skills, thus encouraging students to leave the rural community (DeYoung, 1990).
It is unclear what educational programs--academic or occupational--contribute
more to fostering community attachment among rural youth. Using HS&B data,
it is possible to trace school curriculum offerings and student course-taking
and to link these measures to students' post-school mobility. Importantly,
HS&B provides local labor market indicators that are critical factors for
PRACTICAL SUGGESTIONS FOR USING NCES RESOURCES
locale differentiation. The definition of rural continues to evoke discussion in
rural education circles. While all NCES datasets provide measures of locale,
some involve refined categorization, such as the NCES locale scheme that takes
into account population density and size and distance to metropolitan centers
(used in SASS). Others have simplistic distinctions such as a dichotomy of
metropolitan versus nonmetropolitan areas (e.g., NHES). Analysts should make
sure that the dataset contains a suitable locale measure.
CD-ROM and related techniques. CD-ROM technology stores vast amounts of
information on small disks to provide easy access. NCES's electronic code books
enable quick review and extraction of variables contained in different data
files. It is a great advantage to have a CD-ROM drive on your computer.
Getting what you need from the bureaucracy. The National Data Resource
Center, sponsored by NCES, provides customized data sets and tabulations as
requested, free of charge. The contact person is Carl Schmitt (202/219-1642).
You can also access the U.S. Department of Education's Internet Gopher server
(gopher.ed.gov) or World Wide Web homepage (http://www.ed.gov) to review NCES
data products and access some data files; or send your request via E-mail
(firstname.lastname@example.org). Other suggestions:
Call the Data Resource Center to sign onto its mailing list to get updated
information about NCES projects.
Define your data needs as precisely as possible: If your needs are specific and
straightforward, simply call the Data Resource Center and make your requests; if
you need to analyze data yourself, call the specific staff members who know the
most about a particular project and discuss your needs with them.
If you need to use NCES data regularly, attend an NCES seminar.
Try to access data through the telecommunications network, which is becoming
increasingly efficient and easy to use.
Davis, C., & Sonnenberg, B. (1993). Programs
and Plans of the National Center for Education Statistics. Washington, DC:
National Center for Education Statistics. (ED 360 389)
DeYoung, A. J. (1991). Introduction. In A. J. DeYoung (Ed.), Rural education:
issues and practice. (pp. XV-XXI). New York: Garland Publishing.
Haas, T. (1992). Leaving home: Circumstances afflicting rural America during
the last decade and their impact on public education. Peabody Journal of
Education, 67(4), 7-28.
Herriot, R. (1992). School district-level statistics from the decennial
census. (Unpublished Manuscript, National Center for Education Statistics,
Hobbs, D. (1991). Rural education. In C. B. Flora & J. A. Christenson
(Eds.), Rural policies for the 1990s (pp.151-165). Boulder, CO: Westview Press.
Ingels, S. J., Thalji, L., Pulliams, P., Bartot, V. H., Frankel, M. R.
(1994). National Education Longitudinal Study of 1988. Second Follow-Up: School
Component Data File User's Manual. Contractor Report. Washington, DC: National
Center for Education Statistics. (ED 376 212)
Stephens, E. R. (1992). Mapping the research task for the construction of a
federal system for classifying the nation's rural school districts. Journal of
Research in Rural Education, 8(3), 3-28.
Stern, J. D. (Ed.) (1994). The condition of education in rural schools.
Washington, DC: U.S. Department of Education, Office of Educational Research and
Improvement. (ED 371 935)
Theobald, P. (1992). Rural philosophy for education: Wendell Berry's
tradition. Charleston, WV: ERIC Clearinghouse on Rural Education and Small
Schools. (ED 345 930)
Wigginton, E. (1985). Sometimes a shining moment. Garden City, NY: Anchor