[R] ANOVA 1 too few degrees of freedom

peter dalgaard pdalgd at gmail.com
Fri May 6 08:31:55 CEST 2011


On May 5, 2011, at 23:30 , Rovinpiper wrote:

> Thanks slre,
> 
> I seem to be making some progress now.
> 
> Using a colon instead of an asterisk in the code really changes things. I
> had been getting residual SS and MS of zero. Which is ridiculous. Now I get
> much more plausible values. 
> 
> Also, When I used an asterisk instead of a colon It wouldn't give results
> for three way interactions. With colons it will.
> 
> You are correct about plot being nested within treatment. There are six
> plots in each of 2 treatments.
> 
> So, I guess I will have to perform a separate analysis to quantify the
> effect of treatment.
> 

Beware that as you have highly significant effects of plot and its interaction with day, and plot being nested in treatment, you can't test for treatment or treatment:day effect with a systematic effect of plot  and plot:treatment in the model (you are only getting p values because of the sequential computation of the anova table - if you put plot before treatment, you'd get zero df). More likely, you want to make the "plot" terms random, as in ~treatment*day + Error(plot/day)


> Thanks again.
> 
> Analysis of Variance Table
> 
> Response: Combined.Rs
> 
> Df  Sum Sq   Mean Sq   F value          Pr(>F)    
> Combined.Trt                                                             1  
> 52.80     52.805     96.2601       < 2.2e-16 ***
> Combined.Plot                                                            10 
> 677.69   67.769     123.5380    < 2.2e-16 ***
> as.factor(Combined.Day)                                            16 
> 2817.47 176.092   321.0041    < 2.2e-16 ***
> Combined.Trt:as.factor(Combined.Day)                      16   47.82    
> 2.989       5.4487        4.048e-10 ***
> Combined.Trt:Combined.Plot:as.factor(Combined.Day)80  455.42   5.693      
> 10.3776      < 2.2e-16 ***
> Residuals                                                                   
> 284 155.79   0.549                       
> ---
> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 
> 
> 
> --
> View this message in context: http://r.789695.n4.nabble.com/ANOVA-1-too-few-degrees-of-freedom-tp3493349p3499649.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> 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.

-- 
Peter Dalgaard
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com



More information about the R-help mailing list