www.cmm.bristol.ac.uk/forum

hi everybody

I have a TIMSS data and I am interesting on teachers answers...I am looking on threelevel country, class and teacher ....
so when I am develop the equations I got for examplw these ASMM01= b*consijk+ATB15ijk ....I thought that I must get ATB15jk because teachers answer...

I recode the data and I think that now each ID on country class and student level is unique...Is it something else that maybe cause this problem??so I can check it again..

Thank you very much for the help
Katerina

The data also has to be sorted according to the model hierarchy, so you could check whether this is the case. You can also check what MLwiN thinks the hierarchy structure currently is by opening the Model->Hierarchy viewer window.

Dear Chris
I sorted my data but I still have this problem....What should I have to check in hierarchy viewr window to see whether there is problem...

Thanks in advance for your help
Katerina

You should check that the number of units and their sizes are what you expect them to be at each level. If they are then your IDs should be correctly specified.

Hi Chris

ok I have checked it and all data are ok....I cannot understand why in some countries the levels work correct but in others dont.
Are there anything else I should check or do.....

Thank you in advance again
Katerina

Dear Chris

At last I found where was the problem..

Thanks for your advices
katerina

I'm glad to hear that you have tracked down the issue. Is there any chance you could say how you fixed it, in case someone has a similar problem in the future?