You are here

OpenMx status is 6

4 posts / 0 new
Last post
Sepidehe's picture
Offline
Joined: 05/24/2019 - 11:00
OpenMx status is 6

Dear Mike and colleagues,
I am trying to conduct a metaSEM analysis with 11 variables and 111 studies. However, I am stuck at the early stages as the OpenMX status is 6. Due to this issue, I removed all the studies with only one input (i.e., corrected correlation for unreliability) and performed the analysis again but the problem still persists. I am attaching my data as well as the results of my analyses. I would appreciate if you could help me figure this issue out.
Many thanks,
Sepideh

Mike Cheung's picture
Offline
Joined: 10/08/2009 - 22:37
Dear Sepideh,

Dear Sepideh,

The error is due to the issue of reading missing data as 0. After correcting it, it works okay. However, you may need to be careful because the numbers of studies for the average correlations are quite small.

Best,
Mike

File attachments: 
Sepidehe's picture
Offline
Joined: 05/24/2019 - 11:00
Thank you!

Dear Mike,
Thank you so much for finding my (strange) mistake! As for the number of studies, do I still need to worry if I perform power analysis (for both the effect sizes reported in the original studies as well as my hypothesized relationships) and obtain results higher than 0.8?
I really appreciate your helpfulness.
Best,
Sepideh

Mike Cheung's picture
Offline
Joined: 10/08/2009 - 22:37
Hi Sepideh,

Hi Sepideh,

Small number of studies is always a concern in a meta-analysis. If you want to generalize your findings, your studies should be representative to the other studies.

Best,
Mike