|
|
Path modelling of complex data |
|
Message/Author |
|
Joe Boden posted on Thursday, March 23, 2017 - 6:50 pm
|
|
|
Hello I am attempting to fit a model using complex data drawn from a sample of mothers and fathers. What I would like to do is to: a) compare the fit of the model for mothers with that of fathers; and b) account for the fact that mothers and fathers are usually (but not completely) nested within families. My syntax is as follows: GROUPING IS gender (1,2); CLUSTER IS id; ANALYSIS: TYPE = TWOLEVEL; MODEL: parstrt4 ON group (1) hadtott4 (2) socsupt4 (3) dastott4 (4) stot (5) hatot (6); hadtott4 WITH socsupt4 dastott4 stot hatot; dastott4 WITH socsupt4 stot hatot; socsupt4 WITH stot hatot; stot WITH hatot; OUTPUT: sampstat mod(5) stdyx; When I run this model I get an error as follows: *** ERROR Cluster ID cannot appear in more than one group. Problem with cluster ID: 1001 Thanks in advance! |
|
|
The groups must contain independent observations. Members of the same cluster cannot appear in both groups. See Web Note 16 for further information. |
|
Back to top |
|
|