EFFECTIVE DATE: 03/16/87   CES STANDARD 87-03-03
PURPOSE: To ensure that studies routinely perform those actions necessary to study the causes of and to minimize both survey and item nonresponse.      
    o Nonresponse (or total nonresponse) refers to the failure of the survey respondent, or for administrative record studies, of the person or organization providing the data, to provide any information in response to the data request.   Item nonresponse refers to certain items in the data request being missing when initially a response is expected.      
    o For total nonresponse, in studies conducted by the center or its contractors, very serious efforts should be made to collect information not obtained initially in the survey.   Callbacks with the respondents or data providers will be scheduled during and immediately after the data collection period. Records on the number of callback attempts and their outcome should be kept, and a final disposition of the case (the reason it continues as a nonresponse case) should be kept in lieu of a completed interview or shuttle form. Attempts should be made to determine characteristics of the nonrespondents for use in weighting and imputation.
  o   For item nonresponse, key statistics to be collected will be identified in the analysis plan; some provision should be made in the field operations to monitor the responses on these items to ensure that no one interviewer (in a survey) has an exceptionally high item nonresponse rate for key items.   The project officer should also consider a field or office edit that would require telephone followup for cases with key items missing.    
  o   The survey design should consider explicitly the motivation of respondents, and the kinds of messages and publicity that might improve total response.      
  o   In addition, well-designed survey or shuttle forms and well-trained interviewers or clearly written survey (shuttle) instructions will be the most efficient devices for ensuring a high response. Procedures for minimizing nonresponse must be well documented. These procedures should be reviewed as part of and in conjuction with any pretest of the survey questionnaire or the use of shuttle forms for administrative record data collection.  
RELATED STANDARDS:                  
      o CES Standard for Planning Statistical Surveys, 87-01-01          
      o CES Standard for Testing Data Collection Instruments, 87-03-01 
o CES Standard for Treatment of Item Nonresponse, 87-05-01