[R] ANOVA Estimated effects may be unbalanced

Bert Gunter gunter.berton at gene.com
Mon May 20 04:23:20 CEST 2013

I would say  that the OP should seek local statistical help, as he
appears to be out of his statistical depth. This would appear to be a
mixed effects models-type setup, but a local statistical expert would
be much better able to judge what the goals of the study were and what
sort of approach, including graphics, might be appropriate.

Seeking remote statistical help from this list when your statistical
background is weak is

a) Inappropriate -- this is not a statistical help list; it's for R software;

b) Highly risky, as relevant details  of the underlying context and
the way the data were obtained are likely to be overlooked. It is
**not** merely a question of finding the appropriate function and just
turning the crank, although that's the way many seems to approach data

On Sun, May 19, 2013 at 6:00 PM, Jim Lemon <jim at bitwrit.com.au> wrote:
> On 05/20/2013 10:52 AM, Luis Fernando García Hernández wrote:
>> Dear All,
>> This is a data relating leg shaking on differenntre treatments. I reused
>> several individuals so I want to know 1) if there are significative
>> differences on shaking per treatment and 2) if the reused individuals
>> presented some effect or significative variation.
>> Nevertehless when I make
>> model1<-aov(Legshaking~Idmale*Idfemale*tratament)
>> or
>> model1<-aov(Legshaking~Idmale+Idfemale+tratament)
>> I get
>> 13744 out of 13824 effects not estimable
>> Estimated effects may be unbalanced
>> What could be wrong here? Any advice about how to handle this data in R
>> will be welcome!
> Hi Luis,
> Are you sure that you don't want to compare males and females as a group and
> not every individual male and female?
> Jim
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.


Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374

More information about the R-help mailing list