

Multiple Group Cohort Estimated Time ... 

Message/Author 


Using panel data from a largescaled longitudinal survey, we aim to investigate how personality changes during to, prior and after marriage. Personality was assessed at 3 time points, in 2005, 2009 and 2013. Marriage was assessed in yearly intervals. We considered all subjects with marriage from 2001 to 2016 (N=3677) and applied multiple group multiple cohort growth modeling to combine their data: They married in different years and thus, time spans between marriage and personality assessment vary. Participants with marriage in the same year were considered as one group, resulting in 16 groups. We would now like to freely estimate the developmental trajectory. Our question is: Is it possible to use freely estimated time scores in this model and when yes, how can we do so? We did not find any information on this online. Also, we get the following warning (for each group): WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IN GROUP 16 IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/ RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES. CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLES. How can we solve this? The program created a diagram, but when we try to open it, an empty page appears. Thank you very much for your help and time! 


Q1: Yes, just hold free time scores equal across cohorts/groups when they represent the same time. This can be done using the BY approach for the slope growth factor instead of using the bar (  ) approach (see UG V8 pages 747752). Q2: Did you check TECH4? 


Thank you! I started with your online example and tried to rewrite the syntax by using the BY approach. Original: MODEL: i s y1@0 y2@.2 y3@.4 y4@.6; [i] (1); [s] (2); i (3); s (4); i WITH s (5); y2y4 (2224); MODEL 1989: i s y1@.1 y2@.3 y3@.5 y4@.7; y1y4; Modified: MODEL: i BY y1y4@1; s BY y1@0 y2@.2 y3@.4 y4@.6; [y1y4@0 i s]; [i] (1); [s] (2); i (3); s (4); i WITH s (5); y2y4 (2224); MODEL 1989: i BY y1y4@1; s BY y1@.1 y2@.3 y3@.5 y4@.7; y1y4; My results are different and obviously wrong, but I failed to solve this issue. Could you give me an example syntax how your original needs to be written correctly with the BY approach (fixed time scores)? And how it needs to be modified to use free time scores? Many thanks! Eva 


This should work. First make sure that the 2 runs have the same number of parameters. If they do, send both outputs to Support along with your license number. If they don't, find out why by looking at TECH1 or the estimates in the output. 

Back to top 

