|
Message/Author |
|
|
Dear Drs. Muthen, I was wondering if there is a way to conduct an analysis in Mplus parallel to the one described by Judd, Westfall, and Kenny (2012) DOI: 10.1037/a0028347. They describe a model for within-subjects designs that involve participants responding to a series of stimuli. They point out that the data are clustered not only within subjects, but also within stimuli. However, the clusters overlap--it's not like students nested in classrooms. Here's the SPSS syntax, where i is subject and i is stimulus: MIXED y WITH c /FIXED=c /PRINT=SOLUTION TESTCOV /RANDOM=INTERCEPT c | SUBJECT(i) COVTYPE(UN) /RANDOM=INTERCEPT | SUBJECT(j). Thank you! |
|
|
Check out Type=Crossclassified examples in the UG. There is also a video and handout about this in our Short Course Topic 10 on our website. |
|
Back to top |
|
|