CENTER FOR EDUCATION STATISTICS
OFFICE OF EDUCATIONAL RESEARCH AND IMPROVEMENT 
U.S. DEPARTMENT OF EDUCATION


                                     
SUBJECT: STANDARD FOR TESTING DATA COLLECTION INSTRUMENTS AND DATA COLLECTION SYSTEMS          
                       
                                     
EFFECTIVE DATE: 02/09/87    

CES STANDARD 87-03-01

       
                                     
  PURPOSE: To ensure that the data instruments and systems used by CES to collect data are adequately tested before a survey is conducted.              
                                     
    o The Project Officer must review questionnaires to determine that each question is needed for the survey objectives and that overall the questions will provide the needed data.    
                                     
    o Questionnaires developed for all CES surveys must be pretested for feasibility before field testing by the Project Officer, other CES staff and representatives of persons and groups who will participate in the survey.  
      The questionnaire should be reviewed for clarity, flow of questions, adequacy of instructions and length of response time, appropriate grouping and sequencing of questions.
                                     
    o once a questionnaire is developed it must be pretested in the field. Sample size for the pretest should be sufficiently large to identify potential problems and therefore should not be limited to fewer than ten to avoid OMB clearance procedures.   In most cases, purposive sample designs for a pretest are preferable to probability samples in order to ensure that important subpopulations are included.  
                                     
    o Provision for successive pretests should be made when the first indicates the need for a substantial change in the questionnaire or the system.
                                     
    o The method or system by which the data will be collected must also be tested. For face-to-face and pre-arranged telephone interviews this includes procedures for contacting respondents, setting up meeting times and places, notifying respondents of materials they may need to review or have at the interview, and other logistical considerations. For mailed questionnaires, procedures for sending and receiving the questionnaire, logging responses and follow-through on non-response must be tested.
                                     
    o For surveys based on an analysis of administrative records, a field-based test of the survey collection instrument must be done to ensure that the information requested is available without undue data burden; that definitions and instructions are provided and are comparable among the reporting units; and that appropriate staff to respond are identified.      
                                     
    o Quality control procedures should be tested before beginning a major survey. Quality control procedures may include validity checking, data-editing, followup on non-response and data processing checks.  
                                     
  RELATED STANDARDS:                              
                                     
    o CES Standard for Planning Statistical Surveys, 87-01-01 
o CES Standard for RFP Development for Surveys, 87-02-01
                 
                                     
  RELATED CHECKLISTS, FORMS AND DOCUMENTS                      
    o Testing Data Collection Instruments and Data Collection Systems Checklist.