Author Topic: ConQuest freezes when trying complex between-item multidimensional model  (Read 529 times)

gfulmer

  • Guest
Hello,

I am working with some colleauges who are implementing a questionnaire with 32 rating scale items, consisting of 8 expected dimensions with 4 items.  I've been able to run a unidimensional model without problem, but trying to compare those results with the between-item multidimensional model is giving me trouble.  Specifically, I try to run that latter model, Conquest pops up an error dialog (without any text listed), and then ConQuest appears to be frozen. Most commands do not work; even choosing "Stop" from the "Run" menu does not stop it. I have to quit and re-start ConQuest to get it going again.  This happens every time I run this model.

I'll attach a screenshot that shows this "blank" error message.

I would greatly appreciate any advice for troubleshooting this issue.

ConQuest

  • Global Moderator
  • Newbie
  • *****
  • Posts: 24
    • View Profile
    • Email
Re: ConQuest freezes when trying complex between-item multidimensional model
« Reply #1 on: October 23, 2012, 01:26:37 AM »
You are using an 8-dimensional model with default estimation.  That means 15 to the power 8 nodes.  My apologies that it is not reporting a more informative error message, I'll fix that.

What you should be doing is using montecarlo estimation, not gauss-hermite quadrature.

gfulmer

  • Guest
Re: ConQuest freezes when trying complex between-item multidimensional model
« Reply #2 on: October 23, 2012, 01:29:05 AM »
OK, got it. Thanks! Let me give it a try.

gfulmer

  • Guest
Re: ConQuest freezes when trying complex between-item multidimensional model
« Reply #3 on: October 23, 2012, 01:36:01 AM »
You are using an 8-dimensional model with default estimation.  That means 15 to the power 8 nodes.  My apologies that it is not reporting a more informative error message, I'll fix that.

What you should be doing is using montecarlo estimation, not gauss-hermite quadrature.

Yes, it runs! I re-read the manual section on multi-dimensional models and it did suggest to use Monte Carlo for higher dimensions.