[R] error MANOVA in R

[peter dalgaard]

> On 30 Mar 2015, at 17:11 , Gian Maria Niccolò Benucci <gian.benucci at gmail.com> wrote:
> 
> Dear R-usrs,
> 
> I am trying to perform a MANOVA on a data frame with 31 columns about soil
> parameters and 1 column containing the explanatory variable (Fraction) that
> have three levels.
> my code is the following:
> 
> datam <- read.table("data_manova2.csv", header=T, sep=",")
> names(datam)
> 
> manova_fraction2 <- manova(cbind(pH, AWC, WEOC, WEN, C.mic, CO2.C, Ca, Mg,
> K, Na, sol.exch.Fe, easily.reducible.Fe, amourphou.Fe.oxide...Fe.OM,
> crystalline.Fe.oxides, TN, TOC, NH4.N, NO3.N, N.org, organic.P, avaiable.P,
> Total.PLFA, Tot.Bat, Gram., Gram..1, Funghi, AMF, protozoa, actinomiceti,
> non.specifici) ~ as.factor(Fraction), data= datam)
> 
> summary(manova_fraction2)
> 
> when I did the summary I got this error
> 
>> summary(manova_fraction2)
> Error in summary.manova(manova_fraction2) : residuals have rank 18 < 30
> 
> ​Is this error possibly due to high correlation between my variables?
> 

Nope. Too few observations. For classical MANOVA, you need  > p degrees of freedom to determine the covariance matrix with full rank. As far as I can tell, you have only 21 observations (21-3=18 degrees of freedom) for your p=30 response variables. 

> Many thanks in advance,

-- 
Peter Dalgaard, Professor,
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