[R] boxplot

[R. Michael Weylandt]

A few things in play here:

1) I'm guessing you are new to R, so I'd advise you to take some time
to read some introductory materials at this point. If you type
help.start() into your R session, a good introductory manual will be
available.

2) You don't need any of this "textConnection" business, that's just
one way of reading in data from an email. If you are reading in your
data via read.table() you don't need anything else.

3) Your code should perhaps look something like this, but I can't be
certain without an actual sample of your data:

lista <- read.table("lista.txt",header=T,dec=",")
vit<-subset(lista, (lista$familia == "Vittariaceae") ,
select=c(ingreso, familia,
genero,categoria,ira5,ira6,ira7,ira8,ira9,ira10,irah5,irah6,irah7,irah8,irah9,irah10,ed5))
boxplot( t(vit[grepl("ira",rownames(vit)),]), col =
ifelse(vit[,NCOL(vit)] == "C", 2,3))

but at this point, I'd spend some a little more time learning how to
manipulate and subset in R before trying to handle this finer sort of
detail. I'll quickly parse that last line for you, though, just in
case this is a time sensitive project:

working from the inside out:

grepl("ira", rownames(vit)) # Identify those rows that have ira in the name
vit[grepl("ira", rownames(vit)),] # Select only those rows from vit
t( ) # transpose since boxplot takes in columns (see my above note)

vit[,NCOL(vit)] # isolate the last column
 ** == "C" # Test for being of type "C"
ifelse( **, 2, 3) ## assign either a two or a three depending whethere
it was a "C" or not
col = # color the boxplots

4) On another note, are you sure you want to do your boxplots rowwise?
It's pretty standard in R that columns represent similar data while
rows are independent measurements/observations.

5) And just for the record, the error you got is because you passed
boxplot something that wasn't a data object, but rather an empty
textConnection().

Hope this helps,

Michael

PS -- It's good form (and usually more helpful to you) to cc the whole
list on each step of the correspondance. It lets other voices chime in
if I start leading you astray as well as ensuring that there are
replies coming even at all hours of the day since R-help has a
world-wide readership. Also, it makes sure everything gets archived
nicely so someone can read them in the future.


On Fri, Oct 14, 2011 at 9:19 AM, Ruth Arias <rueu_ac at yahoo.es> wrote:
> thanks for answer me michel
>
> I made some modifications to your suggestion.
> I have tried this:
>> lista <- read.table("lista.txt",header=T,dec=",")
>> vit<-subset(lista, (lista$familia == "Vittariaceae") , select=c(ingreso,
>> familia,
>> genero,categoria,ira5,ira6,ira7,ira8,ira9,ira10,irah5,irah6,irah7,irah8,irah9,irah10,ed5))
>> vitbp<-textConnection("vit")
>> closeAllConnections()
>> boxplot(vitbp[,1:5], col = ifelse(V[,6] == "C", 2,3))
> and I have this answer:
> Error en plot.window(xlim = xlim, ylim = ylim, log = log, yaxs = pars$yaxs)
> :
>   se necesitan valores finitos de 'ylim' = needs finite values
> Además: Mensajes de aviso perdidos = also: message
> 1: In min(x) : ningún argumento finito para min; retornando Inf = no finite
> argument for minimun
> 2: In max(x) : ningun argumento finito para max; retornando -Inf = no finite
> argument for maximun
>
> and what I want is to make a box plot to let me see the differences in
> different iras (ira5, ira6 ...) between the category C and E
>
>
>
> I'm not sure what the "column to identify other the other columns"
> maps to graphically, but perhaps something like this will get you
> started
>
> V <- read.table(textConnection("
> 4        5        6        7        8        site
> 23    56        41      45    63        C
>
> 21    89        42      10  63        E
>
> 32    45        14      17    96        E
>
> 45    74        13      63    41        C
>
> 68    32        10      20    03        E
>
> 95    10        84    45    96        C
> "),header=TRUE)
> closeAllConnections()
>
> boxplot(V[,1:5], col = ifelse(V[,6] == "C", 2,3))
>
>
> Michael
>
> On Thu, Oct 13, 2011 at 5:54 PM, Ruth Arias <rueu_ac at yahoo.es> wrote:
>> hello
>>
>> I want to make a boxplot with diferents rows and also include a column to
>> sort into two groups to each of the other columns
>>
>> my  date set looks like this:
>>
>> 4        5        6        7        8        site
>> 23    56        41      45    63         C
>>
>> 21    89        42      10   63        E
>>
>> 32    45        14      17    96        E
>>
>> 45    74        13      63    41        C
>>
>> 68    32        10      20    03        E
>>
>> 95    10        84    45    96        C
>>
>> THANKS
>>
>>        [[alternative HTML version deleted]]
>>
>>
>> ______________________________________________
>> 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.
>>
>>
>
>
>