|
|
| Message/Author |
|
|
|
|
| Hi, I have a longitudinal dataset. I have been running SEM on the data. The data I have is from twins which has ID for twin 1 and twin 2. I wanted to run a phenotypic SEM (not genetic) but having the ID for both twins clustered. I am not sure how this can be done. I also do not want to lose half of the dataset as this will reduce the power. |
|
|
|
|
You have two options. If the data are in long format with a cluster variable for the twin pairs, you can do TYPE=COMPLEX with CLUSTER = pair;
If the data are in wide format with variables for twin 1 and variables for twin2, non-independence of observations is taken into account by multivariate modeling of the data. |
|
|
| Back to top |
|
|
