Interpretation multilevel


New Member

Ik have a question about the interpretation of the estimates of fixed effects table in a multilevel analyses. In the analysis I compare 5 countrycluster (reference = cluster 1). The clusters are recoded in a manner that in the output cluster1=0 actually means cluster1=1. From the following output I can conclude that only cluster 2 and 4 are significant different from cluster 1 (p<.05). My question is, can I use the estimates of the insignificant clusters also to create a ranking about which clusters scores the best on the output variable? The cluster that scores the best would be then cluster 2 (followed, by cluster 3, 5, 1, 4)?



Last edited: