Question about CCMM-GCM

[auctor21]

Hello,
I am Seong Hun Kim from South Korea.

I write this email hoping to solve a few problems that occurred while using the MLwiN program.

It was an error that occurred while analyzing CCMM-GCM using the MLwiN.
While looking at the paper “When People Move: Using Cross-Classified and Multiple Membership Growth Curve Modeling in Non-Hierarchical Multilevel Data Structures” presented by Bess Rose in AEA conference 2013, I too decided to use the MLwiN to analyze the same CCMM-GCM data.

Especially, after sorting the data and setting the naive model, I tried to set the N classification of the Y variable as 4, expecting the classification to show 4,3,2,1 respectively. However, it did not. When the N Classification is set as 1,2,3,5 the expected results show up, but whenever the N classification is set as 4, the program does not run. Why might this be happening?

Besides this, if there are any other guidelines or manual regarding the CCMM-GCM(Cross-Classified and Multiple Membership Growth Curve Modeling) analysis, I would really appreciate it if you could send a copy of those. For Cross-Classified data, there is a guide book “Module 12_MLwiN Practical Cross-Classified Multilevel Models” and for Multiple Membership, there is “Module 13_MLwiN Practical Multiple Membership Multilevel Models” but for CCMM-GCM, there is no provided guideline, limiting the amount of analysis that can be carried out.
I sincerely request you.

Thank you.

[ChrisCharlton]

I think that this problem may be related to a bug in the current version of MLwiN where it will crash if you try to specify the number of levels as maximum number allowed minus 1. Could you try increasing the maximum number of levels via Options->Worksheet to see whether this fixes it?