Discussion: View Thread

Expand all | Collapse all

AW: [OB-LIST] Proxy items measuring household income/wealth

  • 1.  AW: [OB-LIST] Proxy items measuring household income/wealth

    Posted 09-12-2012 07:58

    Hi Tim,

     

    simply using any „proxy" variable for a construct of interest is dubious in my point of view. If you have evidence of the correlation between such a item and the true income, you could

    create a latent measurement model in which you fix the error variance of the proxy-variable (as an indicator of the latent income variable) to 1-r^2.

     

    In this model, you'd have a latent/unobserved income-variable and perhaps other/dependent variables which can be regressed on the income variable, resulting in an unbiased

    estimate of the relationship between income and the dependent variable(s).

     

    In case you have no correlation between the proxy and income, even better than using the proxy as it is would be to specify a reasonable measurement error and to use this in a fixed

    single-indicator model. Using the proxy as it is has the implication that you assume a correlation between the proxy and income of 1 – which I hope you agree is unreasonable.

     

    Best,

    Holger

     

     

    ___________________________________________________
    Dr. Holger Steinmetz
    University of Giessen

    Faculty of Economics and Business Administration

    Department of Human Resource Management, Small Business Enterprises, and Entrepreneurship

    Licher Str. 62

    35394 Giessen

    Phone: +49 641 99 22103

    Fax: +49 641 99 22109

     

    Holger.Steinmetz@psychol.uni-giessen.de

    http://wiwi.uni-giessen.de/home/personal

    ___________________________________________________
     

     

     

     

     

    Von: Organizational Behavior Division Listserv [mailto:OB@AOMLISTS.PACE.EDU] Im Auftrag von Timothy Thorley
    Gesendet: Dienstag, 11. September 2012 17:15
    An: OB@AOMLISTS.PACE.EDU
    Betreff: [OB-LIST] Proxy items measuring household income/wealth

     

    Hello,

    I am trying to come up with survey items appropriate for high school age subjects that act as a proxy for parent/household income. Examples might include: "How many bedrooms are in your home?" or "How many cars does your family have?" I would appreciate any suggestions or examples of questions(with references if possible) you have found to be effective. Thanks.

    --

    Tim Thorley

    Ph.D. Student
    Department of Management
    VCU School of Business
    301 W. Main Street, Box 844000
    Richmond, VA 23284-4000
    E-mail: thorleyte@vcu.edu
    Phone: 435 773 5760