Discussion: View Thread

  • 1.  Between group comparisons using AMOS Structural Equations Modelling

    Posted 02-28-2007 20:04
    I'm attempting to start using SEM with AMOS, and need to compare the significance levels of goodness of fit of a model across several groups. The model has 10 latent variables from a 57 item survey.

    I've been reading extensively and can find no description that I can understand defining how to do between group comparisons.

    If you can give me a starting place within AMOS or recommend a textbook, article, or link dealing with the topic I will be most appreciative.

    This answer will probably not be of general interest so you may reply directly to me, littrellaom@yahoo.co.nz

    The AoM research methods SIG apparently doesn't have a list server.

    Thanks,
    Romie


    "International business is the general case; local business is the special case." - Peter J. Buckley
    Romie F. Littrell, BA, MBA,PhD, FIAIR, An fánaí fiáin
    AUT Business School N.Z., romie.littrell@aut.ac.nz
    http://www.romielittrellpubs.homestead.com/
    http://www.crossculturalcentre.homestead.com/
    Facilitator, Leadership & Management in Sub-Sahara Africa Conferences

    The all-new Yahoo! Mail goes wherever you go - free your email address from your Internet provider.


  • 2.  Between group comparisons using AMOS Structural Equations Modelling

    Posted 03-01-2007 01:06
    Romie,

    You may want to try a search of the SEMNET listserv archives at
    http://alabamamaps.ua.edu/archives/semnet.html. The listserv has moved
    to http://www2.gsu.edu/~mkteer/semnet.html, which also has a searchable
    archive. I would offer more direct assistance, but I use LISREL and
    wouldn't be able to help with AMOS-specific questions.

    Good luck!

    Barry Stennett

    On Feb 28, 2007, at 8:04 PM, Romie Littrell wrote:

    > I'm attempting to start using SEM with AMOS, and need to compare the
    > significance levels of goodness of fit of a model across several
    > groups. The model has 10 latent variables from a 57 item survey.
    >
    > I've been reading extensively and can find no description that I can
    > understand defining how to do between group comparisons.
    >
    > If you can give me a starting place within AMOS or recommend a
    > textbook, article, or link dealing with the topic I will be most
    > appreciative.
    >
    > This answer will probably not be of general interest so you may reply
    > directly to me, littrellaom@yahoo.co.nz
    >
    > The AoM research methods SIG apparently doesn't have a list server.
    >
    > Thanks,
    > Romie
    >
    >
    > "International business is the general case; local business is the
    > special case." - Peter J. Buckley
    > Romie F. Littrell, BA, MBA,PhD, FIAIR, An fánaí fiáin
    > AUT Business School N.Z., romie.littrell@aut.ac.nz
    > http://www.romielittrellpubs.homestead.com/
    > http://www.crossculturalcentre.homestead.com/
    > Facilitator, Leadership & Management in Sub-Sahara Africa Conferences
    >
    > The all-new Yahoo! Mail goes wherever you go - free your email
    > address from your Internet provider.


  • 3.  Between group comparisons using AMOS Structural Equations Modelling

    Posted 03-01-2007 04:49
    Romie the AoM research Methods listserv is rmnet@listserv.unc.edu
    The list is restricted to members so you will need to join the division to access it.

    Can you give a more detailed description of what you are trying to achieve? If you wish to discover whether the same model fits all the groups, then I believe you need to add a constraint to the model which sets path coefficients to be the same in each group. You then compare the overall fit of the model with and without that constraint set. (You first need to establish that your model has reasonable fit in each group). However, I may have misunderstood what you are trying to achieve. Perhaps you could explain the research question you are trying to answer. For example if you have a hypothesis that an effect varies by group then you may be better served by hierarchical linear modelling.

    Alternatively get access to RMNET where you will find people who have more than my 'enthusiastic amateur' level of expertise.

    best regards

    Mark



    Prof. Mark Fenton-O'Creevy, OU Business School



    ________________________________

    From: International Management Division Discussion on behalf of Romie Littrell
    Sent: Thu 01/03/2007 01:04
    To: IMD-L@AOMLISTS.PACE.EDU
    Subject: Between group comparisons using AMOS Structural Equations Modelling


    I'm attempting to start using SEM with AMOS, and need to compare the significance levels of goodness of fit of a model across several groups. The model has 10 latent variables from a 57 item survey.

    I've been reading extensively and can find no description that I can understand defining how to do between group comparisons.

    If you can give me a starting place within AMOS or recommend a textbook, article, or link dealing with the topic I will be most appreciative.

    This answer will probably not be of general interest so you may reply directly to me, littrellaom@yahoo.co.nz

    The AoM research methods SIG apparently doesn't have a list server.

    Thanks,
    Romie


    "International business is the general case; local business is the special case." - Peter J. Buckley
    Romie F. Littrell, BA, MBA,PhD, FIAIR, An fánaí fiáin
    AUT Business School N.Z., romie.littrell@aut.ac.nz
    http://www.romielittrellpubs.homestead.com/
    http://www.crossculturalcentre.homestead.com/
    Facilitator, Leadership & Management in Sub-Sahara Africa Conferences

    ________________________________

    The all-new Yahoo! Mail <http://us.rd.yahoo.com/mail/uk/taglines/default/nowyoucan/free_from_isp/*http://us.rd.yahoo.com/evt=40565/*http://uk.docs.yahoo.com/nowyoucan.html> goes wherever you go - free your email address from your Internet provider.


  • 4.  Between group comparisons using AMOS Structural Equations Modelling

    Posted 03-01-2007 16:20
    I looked at the RM site at aomonline but it didn't mention a list server, or I would have joined yesterday.

    Your suggestion is that a satisfactory technique, after goodness of fit tests, is to fix the parameters on some observed basis (for example, a set of one of the output path diagram coefficents) and run, for exampl, three analysis on each of three samples, and then interpret the results "manually" as to fit. Correct?

    Hierarchical linear modelling is in fact more appropriate. My major problem is that I am encountering "jump on the fad" journal editors and reviewers, who apparently are not well versed in SEM, and reject an article because the analysis does not employ SEM, appropriate or not.

    Thanks for the help.

    Romie


    "M.P.Fenton-OCreevy" <M.P.Fenton-Ocreevy@OPEN.AC.UK> wrote:
    Romie the AoM research Methods listserv is rmnet@listserv.unc.edu
    The list is restricted to members so you will need to join the division to access it.

    Can you give a more detailed description of what you are trying to achieve? If you wish to discover whether the same model fits all the groups, then I believe you need to add a constraint to the model which sets path coefficients to be the same in each group. You then compare the overall fit of the model with and without that constraint set. (You first need to establish that your model has reasonable fit in each group). However, I may have misunderstood what you are trying to achieve. Perhaps you could explain the research question you are trying to answer. For example if you have a hypothesis that an effect varies by group then you may be better served by hierarchical linear modelling.

    Alternatively get access to RMNET where you will find people who have more than my 'enthusiastic amateur' level of expertise.

    best regards

    Mark



    Prof. Mark Fenton-O'Creevy, OU Business School



    ________________________________

    From: International Management Division Discussion on behalf of Romie Littrell
    Sent: Thu 01/03/2007 01:04
    To: IMD-L@AOMLISTS.PACE.EDU
    Subject: Between group comparisons using AMOS Structural Equations Modelling


    I'm attempting to start using SEM with AMOS, and need to compare the significance levels of goodness of fit of a model across several groups. The model has 10 latent variables from a 57 item survey.

    I've been reading extensively and can find no description that I can understand defining how to do between group comparisons.

    If you can give me a starting place within AMOS or recommend a textbook, article, or link dealing with the topic I will be most appreciative.

    This answer will probably not be of general interest so you may reply directly to me, littrellaom@yahoo.co.nz

    The AoM research methods SIG apparently doesn't have a list server.

    Thanks,
    Romie


    "International business is the general case; local business is the special case." - Peter J. Buckley
    Romie F. Littrell, BA, MBA,PhD, FIAIR, An fánaí fiáin
    AUT Business School N.Z., romie.littrell@aut.ac.nz
    http://www.romielittrellpubs.homestead.com/
    http://www.crossculturalcentre.homestead.com/
    Facilitator, Leadership & Management in Sub-Sahara Africa Conferences

    ________________________________

    The all-new Yahoo! Mail <http: mail="" uk="" taglines="" default="" nowyoucan="" free_from_isp="" *http:="" us.rd.yahoo.com="" evt="40565/*http://uk.docs.yahoo.com/nowyoucan.html"> goes wherever you go - free your email address from your Internet provider.
    </http:>



    "International business is the general case; local business is the special case." - Peter J. Buckley
    Romie F. Littrell, BA, MBA,PhD, FIAIR, An fánaí fiáin
    AUT Business School N.Z., romie.littrell@aut.ac.nz
    http://www.romielittrellpubs.homestead.com/
    http://www.crossculturalcentre.homestead.com/
    Facilitator, Leadership & Management in Sub-Sahara Africa Conferences

    New Yahoo! Mail is the ultimate force in competitive emailing. Find out more at the Yahoo! Mail Championships. Plus: play games and win prizes.