0

I have some data from a toxicity assay. My response variable is Plutei, which is the number of normal-shaped larvae (6 replicates), collected:

  • from 2 different species
  • in 5 different times of the year
  • in 3 different sites.

I've treated this larvae with:

  • 3 different metals
  • each one with 6 different concentrations (to analyse the toxicity).

I want to evaluate the significance of each factor (and their interaction) in the Plutei response that I measured, and I wanted to do it with the adonis function in the vegan package, version 2.2-1.

I've created a data frame like this:

Time  Species        Site       Metal  Concentration  Plutei  Replicates
T1    Paracentrotus  Antignano  Zn                60      78           1
T1    Paracentrotus  Antignano  Zn                60      78           2
T1    Paracentrotus  Antignano  Zn                60      77           3
T1    Paracentrotus  Antignano  Zn                60      80           4
T1    Paracentrotus  Antignano  Zn                60      80           5
T1    Paracentrotus  Antignano  Zn                60      79           6
T1    Paracentrotus  Antignano  Zn                70      70           1
T1    Paracentrotus  Antignano  Zn                70      69           2
T1    Paracentrotus  Antignano  Zn                70      70           3
T1    Paracentrotus  Antignano  Zn                70      71           4

This is str(mydata):

str(mydata) 'data.frame': 3241 obs. of 7 variables:
 $ Time : Factor w/ 6 levels "T1","T2","T3",..: 6 1 1
 $ Species : Factor w/ 3 levels "Arbacia","Paracentrotus",..: 3 2
 $ Site : Factor w/ 4 levels "Antignano","Chioma",..: 4 1
 $ Metal : Factor w/ 4 levels "Cd","Cu","Metal",..: 3 4
 $ Concentration: Factor w/ 23 levels "100","100/70",..: 23 15 1
 $ Plutei : Factor w/ 90 levels "0","1","10","11",..: 90 77 77 76
 $ Repl : Factor w/ 7 levels "1","2","3","4",..: 7 1 2 3 4 5 6 1 2 3 ...

When I try to run the adonis function, 

perm <- adonis(Plutei ~ Time * Species * Site * Metal, data = mydata, permutations = 999)

I get this message:

Error in rowSums(x, na.rm = TRUE) : 
  'x' must be an array of at least two dimensions

What am I doing wrong?

Alex A.
  • 5,466
  • 4
  • 26
  • 56
Andrea
  • 1
  • Hello Alex, thank you for your reply. I am using vegan 2.2-1 and this is str(mydata)...as you can see, I can't assign "Plutei" as Y variable, and I don't know how to deal with the replicates – Andrea May 29 '15 at 15:08
  • str(mydata) 'data.frame': 3241 obs. of 7 variables: $ Time : Factor w/ 6 levels "T1","T2","T3",..: 6 1 1 $ Species : Factor w/ 3 levels "Arbacia","Paracentrotus",..: 3 2 $ Site : Factor w/ 4 levels "Antignano","Chioma",..: 4 1 $ Metal : Factor w/ 4 levels "Cd","Cu","Metal",..: 3 4 $ Concentration: Factor w/ 23 levels "100","100/70",..: 23 15 1 $ Plutei : Factor w/ 90 levels "0","1","10","11",..: 90 77 77 76 $ Repl : Factor w/ 7 levels "1","2","3","4",..: 7 1 2 3 4 5 6 1 2 3 ... – Andrea May 29 '15 at 15:09

0 Answers0