[R] Assignment working differently inside ifelse()

[David Winsemius]

On Aug 16, 2011, at 1:10 PM, Stuart Luppescu wrote:

> Hello all, I need to extract rows and columns from a data frame and  
> put
> them in a matrix. In some cases, there are no rows in the data frame
> meeting the selection criteria. For those rows I want to put a row of
> 0's in the matrix. Here's my clumsy code:
>
> tab1.m1 <- matrix(0, nrow=2, ncol=4)
>
> tab1.m1[1,] <- ifelse(length(as.matrix(tab1[tab1$comp==the.comp &  
> tab1$schlid==the.schl & tab1$ext.obs==0, 4:7])) > 0,

> as.matrix(subset(tab1, tab1$comp==the.comp & tab1$schlid==the.schl &  
> tab1$ext.obs==0)[4:7]),
Missing comma about here  
-----------------------------------------------------------------^

>                          0)
> The row of data I want to put in the matrix is:
>
> tab1[tab1$comp==the.comp & tab1$schlid==the.schl & tab1$ext.obs==0,]
>   schlid ext.obs comp          1          2         3         4 NA
> 41  22091       0    1 0.04166667 0.04166667 0.7083333 0.2083333  0
>
> If I do this:
>
> tab1.m1[1,] <- as.matrix(subset(tab1, tab1$comp==the.comp &  
> tab1$schlid==the.schl & tab1$ext.obs==0)[4:7])
>
> I get the correct values in the matrix:
>> tab1.m1[1,] <- as.matrix(subset(tab1, tab1$comp==the.comp &  
>> tab1$schlid==the.schl & tab1$ext.obs==0)[4:7])
>> tab1.m1
>           [,1]       [,2]      [,3]      [,4]
> [1,] 0.04166667 0.04166667 0.7083333 0.2083333
> [2,] 0.00000000 0.00000000 0.0000000 0.0000000
>
> but when I use the ifelse() as above, I get this:
>           [,1]       [,2]       [,3]       [,4]
> [1,] 0.04166667 0.04166667 0.04166667 0.04166667
> [2,] 0.00000000 0.00000000 0.00000000 0.00000000
>
> Can anyone explain this to me?
> Also, is there an easier, less clumsy way to test for the existence  
> of a
> row in a data frame?
>
> Thanks in advance.
> -- 
> Stuart Luppescu -=- slu .at. ccsr.uchicago.edu
> University of Chicago -=- CCSR
> 才文と智奈美の父 -=-    Kernel 2.6.39-gentoo-r3
> You can't expect statistical procedures to rescue
> you from poor data.    -- Berton Gunter (on
> dealing with missing values in a cluster analysis)
> R-help (April 2005)
>
>
>
>
>
> ______________________________________________
> 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.

David Winsemius, MD
West Hartford, CT