www.cmm.bristol.ac.uk/forum

Hi,
I am trying to do a multilevel logistic regression across 28 states in India. So, I have 28 units at second level and around 15128 at the individual level. Some of the independent variable are at the state level. After runnig the model the result show correct subscript , ie, ij for individual level and j for the state level. I have stored the result using store command. The last two rows of the result give the units or the number of observation. The number given for total individual matches but number of units for state does not matches. Instead of 28 it shows 578. I have imported data from spsss for this analysis. Is there any problem with my data?
please please help.
Thank you,
Sandeep

A common cause for the reported number of units to not match what is expected is if the data isn't sorted correctly. See http://www.cmm.bristol.ac.uk/MLwiN/tech ... html#order for further details. You can use Model->Hierarchy viewer to see which units MLwiN has identified.

Dear Chris
Thanks a lot. It was a great help. I was getting very worried about it. Now the results are perfect.
Thanks,
Sandeep