Error while fitting Linear Mixed model

lakshmi_Raj

Hi, 

I have 5 groups of longitudinal (time) data. The sample size is unequal and missing time points. I tried to fit model using mixed model personality. I did Group*time full factorial and had Subject[Group] as my random effect. In the repeated structure, I had Time and subject. But The report shows many values as zeroed or biased and says "confidence level cannot be computed for multiple comparisons with Tukey HSD or students t". where am I going wrong?. 

 

Thanks

Lakshmi

statman

Best guess with the info you provided, is that you do not have enough data to estimate the model you entered  or the model the terms are aliased... Likely not enough DF for the model.  Did you get a singularity error?