I guess the first error occurs because the variable you try to do something (a mathematical operation) is in the wrong format. Sometimes even numvers are stored as characters ("1","2",etc). Try mode() or class () on the variable that causes the trouble.

I think the second error occurs as a consequence of another problem. This error occured to me often, whenever code that transformed data in a way did not work entirely and thus e.g., not enough rows were created to store another variable in.

Can you provide a minimal example of your code in order to verify?

Sven Greving thank you for your response. I am specifically trying to analyze crop data recorded from alpha lattice design. Meta- R is a user friendly software which performs analysis once all information are customized and click on analyze. It is based on R- program but does not require the whole process of commanding like in R- console/ R- studio. Do you perform biological data analysis? I mean crop data analysis. If so, you can help me, I think.

Admittedly, I know nothing about crop data analysis. I however know enough about Meta R to know that it is basically R code with a GUI. Consequently, the error you encounter is not different from the same error when using R directly.

To solve your issue I recommend careful inspection of the data you feed into Meta R. As stated above, Meta R cannot deal with a value you pass in because it is interpreted as something that is not a numerical value. Letters or punctiuation that lingers at the end of a column can cause such problems. One additional idea might be to get R and read the data via

data

Sven Greving This is what I got in the error output,

The following objects are masked from 'package:base':

format.pval, round.POSIXt, trunc.POSIXt, units

Error in Anova.III.lm(mod, error, singular.ok = singular.ok, ...) :

there are aliased coefficients in the model

> if(expdesign=="Augmented") res=try(suppressWarnings(AnaAugmented(expdesign,datos,traits,esite)))

> if(expdesign=="SplitPlot") res=try(suppressWarnings(AnaSplitPlot(expdesign,datos,traits,esite)))

> if(expdesign=="StripPlot") res=try(suppressWarnings(AnaStripPlot(expdesign,datos,traits,esite)))

> if(expdesign=="Factorial") res=try(suppressWarnings(AnaFactorial(expdesign,datos,nfactors,traits,esite)))

> res1=ifelse(inherits(res,'try-error'),"2","1")

> return(list(res1,res))

[[1]]

[1] "2"

[[2]]

[1] "Error in Anova.III.lm(mod, error, singular.ok = singular.ok, ...) : \n there are aliased coefficients in the model\n"

attr(,"class")

[1] "try-error"

attr(,"condition")

Analysis not performed.

Dear Sven Greving,

I tried your suggested method, sapply(data,class) the results was found as

"integer" for most of the observations and "numeric" for one observation. I checked csv file and found no such additional characters like '', associated with data values. But all data observations are separated by comma ( , ). I think that should be valid. What do you have to say? Please suggest any additional ideas.

Keshab Babu Koirala Sir any suggestions? Your suggestions would be much appreciated.

Dear Krishna Dhakal ,

I am somewhat at my wits end here. The error you reported is very distinctive and only occurs when trying to perform mathematical operations with character-variables, I think.

Maybe, you should also check with Meta R-documentation which variables are required in what format, and whether they are correctly selected in the program.

Apart from that, I recommend contacting the programmers of Meta R.

Sorry that I can not be of further assistance and good luck with that issue.

Sven