cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar
Abby
Level I

Lost DFs Error

Hi, Good day all, 

I have an issue running a fit model for my experiment. I have gone through similar problems posted on the platform but I can't seem to pull through the situation.

I have an experiment with 3 main factors [Stations (n = 2), treatment (n = 14) and time (n = 4)] and 10 replicates per sample. When I run a fit model, I get biased, Lost DFs error (kindly see attached).

I am not sure how to go ahead with this error, since it is vital for me to see the effect test and interactions.

I have also attached a sample of my data to see where I may be getting something wrong.

 

 

 

 

3 REPLIES 3
statman
Super User

Re: Lost DFs Error

Welcome to the community.  The lost DF's is usually because you have a model with more degrees of freedom than are available in the data set.  You need to enter the appropriate model for this data and data type (categorical factors).

 

Are you sure the 10 "measures" for each treatment are replicates?  Or are they repeats?

"All models are wrong, some are useful" G.E.P. Box
Abby
Level I

Re: Lost DFs Error

The 10 measures are not repeats but replicates 

Re: Lost DFs Error

I think it is because you are missing a combination. You do not have Time=1 and Treatment=4 data, so the interaction effect is a problem for estimation with least squares regression.

 

dist.PNG