[R] ezAnova difficulties

p@i@di@i m@iii@g oii trusti@d@t@@de p@i@di@i m@iii@g oii trusti@d@t@@de
Wed Jun 12 11:56:27 CEST 2019

I am using ezAnova for the first time and I am a bit lost.

My data "epindata" looks like this:

      studno faculty epin smoker
         1       1    6      2
         2       2   16      1
         3       2   18      2
         4       2   24      1
         5       3   15      2
         6       2   21      2
         .       .   .       .

         .       .   .      .

"studno" ist the number of the observed probant (factor), "faculty" is  
the subject area (1=sciene, 2=philosophy, ..) (factor), "epin" is
the degree of psychological stability (integer), "smoker" means 1=  
yes, 2= no (factor).
The question I want to answer is if smoking and the faculty are  
related with the probants psychological stability.

So what I want to do is a two way independent (!) ANOVA.

Unfortunaly all the examples I found were about data with repeated  
measure and therefore for dependent variables.

I thought my ANOVA should  work like this:

ezANOVA(data=epindata, wid = studno, dv= epin, between= c(faculty,smoker))

It worked but I get this warning which I don't understand:

"Warnung: You have removed one or more Ss from the analysis.  
Refactoring "studno" for ANOVA.
Warnung: Data is unbalanced (unequal N per group). Make sure you  
specified a well-considered value for the type argument to ezANOVA().
Coefficient covariances computed by hccm()"

So I do have some questions


I don't understand the warning . There is no missing value and so the  
groups are equaly long.
And I don't get the "Refactoring" part or what " You have removed one  
or more Ss" means.

Did I get it right that as my variables are independent they both  
belong to "between" , between= c(faculty,smoker)?

I would be really nice if somebody could help me!

Thanks in advance!


More information about the R-help mailing list