Calculate the Satorra-Bentler Scaled Chi-Square using the loglikelihood following an Mplus ML, MLM, MLR or WLSMV see http://www.statmodel.com/chidiff.shtml for more information. Also see a note in Mplus discussion from B. Muthén (e.g., http://www.statmodel.com/discussion/messages/9/189.html#POST7445)
-------------------------------------------------------------------------------
lli , l0(string) p0(string) [ c0(string) ] l1(string) p1(string) [ c1(string) ]
Required commands
l0 - Loglikelihood in analysis model
l1 - Loglikelihood in the unconstrained or less constrained model
p0 - Number of parameters in the analysis model
p1 - Number of parameters in the unconstrained or less constrained model
Options
c0 c1 - Specify the scaling correction factor, default = 1
Example
. lli , l0(-2606) l1(-2583) c0(1.450) c1(1.546) p0(39) p1(47) difference test scaling correction (cd) = 2.014 Chi-square difference test = 22.840 P = 0.004
Author Richard N Jones, ScD Institute for Aging Research Hebrew SeniorLife, Boston, MA jones@mail.hrca.harvard.edu