Specific Multiple correspondence analysis (with passive modalities)

From: Gindo Tampubolon (gindo.tampubolon@manchester.ac.uk)
Date: Wed Mar 30 2005 - 11:26:31 MEST

  • Next message: Ian Martin: "missing value handling & computational precision"

    Dear all,

    My colleague use ADDAD (www.paris.ensam.fr/ADDAD/) to
    conduct specific MCA. Could anyone tell me whether ADE4 can do
    specific multiple correspondence analysis (with passive modalities)?
    Specific MCA deals with the fact that for most social surveys has
    modalities such as `Yes', `No', and ``Don't Know'' or ``Refused to
    answer''. And so, the last two modalities called passive modalities,
    are not included in the construction of the duality diagram (not by
    dropping those observations because those observations might give
    `active modalities': `Yes' or `No' on other questions.)

    I cannot read French but I quote parts from ADDAD manual (p.20)
    that, I guess, relates to this:

    L’analyse des correspondances multiple spécifique (ACM
    spécifique) s’applique à des questionnaires où pour chaque
    question l’individu interrogé ne donne qu’une modalité de
    réponse et unseule, mais où il peut y avoir des réponses
    manquantes, ou des modalités d’effectif très faible, ou des
    modalités de non-intérêt (catégories ‘autres’), etc.

    I'd be most grateful for any help you can give.

    Yours sincerely,

    Dr. Gindo Tampubolon
    Centre for Research on Socio-cultural Change
    University of Manchester

    This archive was generated by hypermail 2b30 : Wed Mar 30 2005 - 10:28:39 MEST