Is it possible to meta-analyze studies that have used mixture modelling to examine the same latent construct? What if the latent construct is measured differently between studies?

EDIT: [These are some additional details, but please let me know if more information is required to answer the question] I am interested in meta-analyzing studies that have examined differential classes of trajectories of cognition in older adulthood. The types of effects are (mostly unstandardized parameters for) baseline cognitive scores, rates of change, and change in rates of decline for each class of individuals. I assume that this process will be complicated due to different studies measuring cognition using different assessments (i.e. the MMSE versus full cognitive batteries that are averaged to yield a composite score of global cognition), and also due to studies deciding on a different number of trajectory classes.

  • 2
    $\begingroup$ I think you are going to have to give more details before anyone can even begin to answer this. $\endgroup$
    – mdewey
    Jan 4 '18 at 18:09
  • 2
    $\begingroup$ A description--even hypothetical--of the types of effects from growth mixture models that you would be interested in synthesizing would probably be helpful. $\endgroup$
    – jsakaluk
    Jan 4 '18 at 18:38
  • $\begingroup$ Well, the short answer is "It probably is possible" but we can't say how without a lot of details. $\endgroup$
    – Peter Flom
    Jan 7 '18 at 13:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.