Statistical Standards
Statistical Standards Program
Table of Contents
1. Development of Concepts and Methods
2. Planning and Design of Surveys
3. Collection of Data
4. Processing and Editing of Data
5. Analysis of Data / Production of Estimates or Projections
6. Establishment of Review Procedures
7. Dissemination of Data

7-1 Machine Readable Products
7-2 Survey Documentation in Reports
7-3 Release and Dissemination of Reports and Data Products

Appendix A
Appendix B
Appendix C
Appendix D
Publication information

For help viewing PDF files, please click here



PURPOSE: To provide the appropriate amount of documentation on the data, methodology, and other important aspects of a survey in each NCES report. Survey documentation in the report should enable the reader-even the non-statistical user-to understand its contents, and the use and limitations of data readily and clearly.

KEY TERMS: coverage, disclosure risk analysis, frame, instrument, key variables, pretest, and probability of selection.

STANDARD 7-2-1: All NCES reports must include documentation that allows the reader to understand the nature and limitations of the results presented. The level of detail included will vary depending on the type of report. The general areas to be covered include: executive summary, status of data, sample design, data collection, and data presentation. The List 7-2-A outlines the types of documentation to be included in the various types of NCES reports. "C" for "Complete" indicates the full item is to be included. "B" for "Brief" indicates that a brief description should be included; and "†" indicates not applicable.

STANDARD 7-2-2: Sampling standard errors must be available for all estimates included in reports. Sampling standard errors (se's) or confidence intervals (CI's) for statistics in tables and graphs can be included in reports in their entirety. In which case, se's or CI's for each table for graph are reported either in a separate table in an appendix, or in columns accompanying the statistics being presented. Alternatively, especially for publications that are targeted to general audiences, a separate table of exemplar standard errors on key statistics may be presented in the technical appendix with the detailed standard error tables for all tables and graphs included in a report available on the Web.

    GUIDELINE 7-2-2A: To caution users who might attempt to independently test for certain differences using the standard errors provided, a cautionary note should be provided with the standard errors, stating the following:

    Some estimates may be correlated with each other. Generating statistical tests for such estimates solely with these standard errors implicitly assumes these covariances are zero and may be different from the actual significance test used in the report.

List 7-2-A. Checklist for documentation
to be included in NCES reports