Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Getting the DIF-corrected self-assessment after estimation with gllamm

    Dear Statalist,

    I have a model with one self-assessment question and two vignettes. Following Rabe-Hesketh and Scrondal (2002), I use the following command:

    Code:
    gen vign = 1-self
    
    order id item i1 i2 self vign xsay_a 
    
    eq het: vign self
    
    eq thresh: $demo 
    
    cons def 1 [lns1]self=0
    gllamm xsay_a  i1 i2  $s_demo, /*
      */ i(id) link(soprobit) s(het) ethresh(thresh) constr(1) /*
      */ init
    This is working perfectly. I now want to get the DIF-corrected self-assessment from the model and I am not sure if I am doing it correctly with:

    Code:
    gllapred lp1, linpred
    Would somebody know if this is correct?

    And a less important question, but maybe of interest to others as well: would the new irt model in Stata14 also be an option here?

    Thank you,
    Mathilde van Dijk

    Reference:
    Rabe-Hesketh, S., & Skrondal, A. (2002). Estimating CHOPIT models in GLLAMM: political efficacy example from King et al.(2002). London: Institute of Psychiatry, King's College.
    Last edited by Mathilde van Dijk; 30 Sep 2016, 06:13.

  • #2
    Dear Mathilde

    I assume you never got a response ?

    I thought it should be

    gllapred pp, mu above(`x')

    but this seems to give very small estimates so I don't think correct.

    ​​​​​​​Declan

    Comment

    Working...
    X