Announcement

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

  • measurment invariance

    Dear STATA users,
    I am experiencing troubles in performing a measurment invariance analysis using the sem command on STATA version 12.1.
    I am using a large dataset (over 15'000 subjects) collected over 8 different sites and about 12 items that load on 4 different factors as shown by a confirmatory factor analysis. I want to check if factor loading is invariant for each individual site. Unfortunately, I do not get any problem as the program iterates very long (still after 4000) and does not converge on any solution. I use following commands:

    model 1, no constrains
    sem (FACTOR1->item1 item2 item3)(FACTOR4->item4 item5 item6)(FACTOR3->item7 item8 item9)(FACTOR4->item10 item11 item12), group(centreid) ginvariant(none)

    model 2, factor loading constrained
    sem (FACTOR1->item1 item2 item3)(FACTOR4->item4 item5 item6)(FACTOR3->item7 item8 item9)(FACTOR4->item10 item11 item12), group(centreid) ginvariant(mcoef)

    Can you give me any advise on how to solve this problem? Is the command right?
    Thanks and sincerely

    Lorenzo

  • #2
    Lorenzo, welcome to Statalist. This forum is for test posts. Please ask your question in the general forum.

    Comment

    Working...
    X