+ Reply to Thread
Results 1 to 2 of 2

Thread: SEM versus multiple regression?

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    SEM versus multiple regression?




    Hello. I have a within-subjects longitudinal design with two time points. There are two DVs and 3 IVs. All are measured at both time points. I used a hierarchical linear regression to look at the effects of the IVs on the DVs, doing a within-time regression (t1 outcomes on t1 predictors/t2 outcomes on t2 predictors) and predicting across time (t2 outcomes on t1 predictors). In the across time regressions, I include the time 1 measure of the outcome variable.

    Someone suggested I should use a cross-lagged SEM rather than regression. My understanding, however, is that multiple regression, using all the variables and controls is simply a special case of SEM and that I am actually getting the same information. Can someone provide some information about how accurate this is and whether there is anything I am missing by using a regression model for analysis?

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: SEM versus multiple regression?


    I have not dealt much with SEM in the time context, only cross sectional designs. There are differences between regression and SEM in that context. For example, SEM gets at indirect effects that regression can not (or not easily anyhow). Also in regression impact only goes one way, X influences Y, Y can not influence X. In some SEM models this is not true, that is it is possible for impact to flow both ways although these are more complex designs. I do not know how that influences time variations.

    Doing a multilevel time influenced regression is brave
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats