COR 690A

INSTRUCTIONS FOR THE INTERVIEWERS REGARDING REPORTING OF CHILDREN 
                   IN THE 1975 GRADUATE SURVEY
                                                
                                
     This memo describes the instructions to the interviewers
concerning questions 117, 118 and 125. Included in this
description are thoughts on the problems involved in the
alternative strategies of cleaning the data on children.  It
should be noted that one aspect of this issue is not discussed
here, that is, the coding of inconsistent responses. When
inconsistent responses to q. 117, 118 and filter in 125 were
found by the coders, these inconsistencies were corrected.

I. Instructions to Interviewers

     This information was obtained from memos sent to the
interviewers and conversations with Bob Lee, field supervisor for
interviewing through the University of Wisconsin Survey Center.  

     The original objective was to include only natural
(biological) children of the respondent. For example, memos to
the interviewers read as follows:

  Q. 117. 'If R asks, number of children include live births, not 
           adopted children or spouse's children from a previous 
           marriage, even if adopted by R.'

  Q. 118. 'You may list adopted as well as biological children, 
           if R gives their names. Please note if child was 
           adopted.'

     According to these original instructions, if children other
than biological ones were mentioned in Q. 117 (and the
interviewer knew that they weren't biological) they were not
included in Q. 117 or Q. 118.  Hence, Q. 125 filter should
reflect biological children only. If 1) a woman had some
biological children and some adopted or stepchildren and 2) only
the biological children were included in Q. 117 and Q. 118, the
respondent was reminded, when answering Q. 125, to consider the
intervals defined by biological children only.

     Some problems arose in the field with this series of
questions, probably because of two reasons. First, the original
instructions could be viewed as ambiguous since Q. 117 specifies
only biological children whereas Q. 118 could include adopted or
stepchildren. Second, the interviewer often found out that
certain children were adopted or stepchildren in Q. 118 or Q.
125m; this implied that non-biological children had been included
in Q. 117. A system was developed to deal with these problems.
The system depended on at what point in the questionnaire the 
respondent revealed that a certain child was adopted or a
stepchild.

   1. If the respondent mentioned in Q. ll7 that some (or all) of 
      his/her children were adopted or stepchildren, then 
      questions 117, 118, 125 should reflect only the biological 
      children of the respondent.

   2. If the respondent first mentioned in Q. 118 that some (or 
      all) of his/her children were adopted or stepchildren, two 
      options were available.

      a. If the respondent did not strongly request that the      
         adopted or stepchildren be included in Q. 118, they were 
         not. Q. 117 was then corrected to reflect biological     
         children only.
      b. If the respondent strongly requested that the adopted or 
         stepchildren be included in Q. 118, they were. Q. 117,   
         118, and 125 would then include biological as well as    
         adopted and stepchildren. In this case Q. 125m should    
         reflect (through a code of 3) that the child was adopted 
         or a stepchild.

   3. If the interviewer did not find out that a child was        
      adopted or a stepchild until Q. 125m, two procedures were  
      followed. Regardless of which procedure was followed, Q.    
      117 and 118 were not corrected to reflect biological        
      children only.

      a. During the first month or two of interviewing, 

         1) If this child was adopted/stepchild and all others 
            were also, the interviewer put not ascertained in the 
            remainder of Q. 125 and proceeded to Q. 126.

         2) If this child was adopted/stepchild and some of the 
            younger children were not, the interviewer tried to 
            get work history information for intervals bound by 
            biological children.

      b.  After the first or second month of interviewing, 
          Q. 125 was continued as if all children were            
             biological. 
          Q. 125m should reflect that the child was               
             adopted/stepchild.