VAR CF0012a PRE CAPI QUESTION SELECTION

VAR CF0012a   PRE CAPI QUESTION SELECTION
              COLUMNS 79-79
              NUMERIC
              MD EQ 9


        QUESTION SELECTION ('FORM') SUMMARY -- PRE
        -------------------------------------------------------------------

        SEE ALSO VCF0012.

        General Note:
        CAPI administration (Computer-Assisted Personal Interviewing) began in
        1996. Similar to the "Form" variable (VCF0012) of previous, non-CAPI
        studies, vars VCF0012A and VCF0012B describe variations in the
        instrument's handling of substantive content-- i.e. variations
        resulting from Study planning for content inclusion. Typically, such
        variations constitute split-sample administration of questions/
        question versions/question placement in order to increase the number
        of survey questions in the study, compare question versions, etc.
        Vars VCF0012/VCF0012A/VCF0012B differ from Cumulative Data File vars
        VCF0015A/B in that the shortened interviews described in the latter
        are the result of problem, error, or adjustments for budgetary or
        administrative necessity, e.g. the demand to meet target for
        completed interviews or to reduce cost.

        Note that with the introduction of CAPI, question content varies no
        longer according to separate printed questionnaires, but according to
        CAPI logic, which presents questions onscreen for the interviewer to
        use.

        In 1996, half-samples were randomly selected to administer each set of
        1996 questions F1-F4 and H4-H8.  Affected variables in the Cumulative
        Data File are VCF0314-VCF0324 (from series F1-F4) and VCF0522 (from
        series H1-H4).

        In 1998, there was no subsampling (e.g. half-samples) in the
        administration.

        In 2000, half-samples were randomly selected to administer E section
        questions and questions F6-F9; respondents not selected for Pre
        administration of section E were assigned to Post administration).
        Affected variables in the Cumulative Data File are VCF0880, VCF0881
        (from E series) and VCF0875, VCF9012 (from series F6-F9).  Respondents
        were also randomly selected to be administered Clinton traits and
        affects in either the Pre or the Post (if not administered, a Clinton
        retrospective was administered instead).  Affected Presidential
        traits/affects variables in the Cumulative Data file are VCF0338,
        VCF0342-349.

        1996:  960035
          0. R administered both F1-F4 and H4-H8 in the Pre
          1. R not administered F1-F4 in the Pre
          2. R not administered H4-H8 in the Pre
          3. R administered neither F1-F4 nor H4-H8 in the Pre

        2000:  000005a,000005b,000005c
          1. R administered both section E and section F in the Pre, with
               Clinton traits and affects
          2. R administered both section E and section F in the Pre, without
               Clinton traits and affects
          3. R administered section E, not administered section F in the Pre,
               with Clinton traits and affects
          4. R administered section E, not administered section F in the Pre,
               without Clinton traits and affects
          5. R not administered section E, administered section F in the Pre,
               with Clinton traits and affects
          6. R not administered section E, administered section F in the Pre,
               without Clinton traits and affects
          7. R not administered either section E nor section F in the Pre,
               with Clinton traits and affects
          8. R not administered either section E nor section F in the Pre,
               without Clinton traits and affects

        ALL OTHER YEARS:
          9. Question not used


==============================