EFFECTIVE DATE: 12/08/86   CES STANDARD 86-04-02
PURPOSE:   To ensure that reports, bulletins, press announcements and data tapes released by the CES accurately report results of surveys and other studies conducted by CES in such a way that the reader can evaluate the conclusions drawn.      
The level of documentation, length and technical nature of the discussion will vary with type of publication and the audiences for whom the publication is intended.  
  o Major reports on surveys prepared by contractors or by CES staff for publication must contain or refer to a technical appendix that discusses all topics required in CES Standard for Survey Documentation. This includes:
    - summary or abstract of when the data were collected, target population and purposes of the survey;          
    - a statement on the status of the data (preliminary, revised or final) ;        
    - a discussion of the methodology used;                
    - a discussion of sampling errors including how the sampling errors were calculated and tables of sampling errors if not in the text;    
    - a discussion of non-sampling errors including sources of error, what was done to control these sources and any external corroborating evidence used to validate the data; and guidelines for using the data.  
  o All key statistics in technical reports should be followed by confidence intervals. Where comparisons are made between- two or more statistics, a statement should be made as to the significance of the test performed
  o Short reports and bulletins must have a statement about the status of the data, sample selection, data collection process, the sampling error for key statistics presented, or alternatively, present confidence intervals after the key statistics. The report must reference more detailed documentation and give the name and phone number of a contact person or office.  
  o Press releases, one or tvo-page bulletins, and responses to ad hoc requests must give the status of the data and include a statement indicating the sampling error for key statistics.      
  o Public release data tapes must conform to CES Standard for Data Tape Preparation including documentation of file structure, record size, count and blocking factors, data element definitions and locations. The data set name (DSN) must refer to the survey name, the series if appropriate, and the version or release date.
  o E.D. TABS and other tabulations prepared in either print, tape or disk format, must discuss status of data, sample design, data collection procedures, salient definitions, response rates, type of data collected, sampling and non-sampling errors.  
  o Proper documentation must be provided for any data reported from a source other than the survey including source name, publication information, table number and page.      
RELATED STANDARDS:                        
  o CES Standard for Survey Documentation, 87-04-01              
  o CES Standard for Codes, Abbreviations and Acronyms, 87-03-05 
o CES Standard for Data Tape Preparation, 87-03-07
  o CES Standard for Estimating Sampling Error, 87-05-02 
o OERI Publication Policy
  o CES Publication Policy                      
  o Maria E. Gonzalez, Jack L. Ogus, Gary Shapiro and Benjamin J. Tepping, "Standards for Discussion and Presentation of Errors in Survey and Census Data," Journal of the American Statistical Association, September, 1975, Volume 70, Number 351, Part II.