Excluding levels in table and xtabs

I'm trying to form contingincy tables among a set of character variables
which were read from a .csv file and
have missing represented as "".  I want to exclude the missing levels
from the table.

levels(CPIC)

[1] ""  "N" "Y"

levels(Manix)

[1] ""  "N" "Y"

xtabs(~CPIC + Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

table(CPIC, Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

The only way I can exclude them is by

t <- table(CPIC, Manix)
t <- t[-1,-1]

that's not to hard in this case, but my application is to a much
larger table where this gets unweildly.

--
Michael Friendly              friendly at yorku.ca
York University               http://www.math.yorku.ca/SCS/friendly.html
Psychology Department
4700 Keele Street             Tel:  (416) 736-5115 x66249
Toronto, Ontario, M3J 1P3     Fax:  (416) 736-5814

I'm trying to form contingincy tables among a set of character variables
which were read from a .csv file and
have missing represented as "".  I want to exclude the missing levels
from the table.

levels(CPIC)

[1] ""  "N" "Y"

levels(Manix)

[1] ""  "N" "Y"

xtabs(~CPIC + Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

table(CPIC, Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

The only way I can exclude them is by

t <- table(CPIC, Manix)
t <- t[-1,-1]

that's not to hard in this case, but my application is to a much
larger table where this gets unweildly.

--
Michael Friendly              friendly at yorku.ca
York University               http://www.math.yorku.ca/SCS/friendly.html
Psychology Department
4700 Keele Street             Tel:  (416) 736-5115 x66249
Toronto, Ontario, M3J 1P3     Fax:  (416) 736-5814

On Wed, 11 Dec 2002, Michael Friendly wrote:

I'm trying to form contingincy tables among a set of character variables
which were read from a .csv file and
have missing represented as "".  I want to exclude the missing levels
from the table.

I think this is a bug. The exclude= argument doesn't work for factors,
because the argument is passed to factor(), and its exclude argument has
a different format when the main argument is a factor. 

        -thomas

levels(CPIC)

[1] ""  "N" "Y"

levels(Manix)

[1] ""  "N" "Y"

xtabs(~CPIC + Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

table(CPIC, Manix, exclude=c("",NA))

    Manix
CPIC        N    Y
     272    4   15
   N 154 2812 1472
   Y 158  466 4870

The only way I can exclude them is by

t <- table(CPIC, Manix)
t <- t[-1,-1]

that's not to hard in this case, but my application is to a much
larger table where this gets unweildly.

--
Michael Friendly              friendly at yorku.ca
York University               http://www.math.yorku.ca/SCS/friendly.html
Psychology Department
4700 Keele Street             Tel:  (416) 736-5115 x66249
Toronto, Ontario, M3J 1P3     Fax:  (416) 736-5814

Thomas Lumley                   Asst. Professor, Biostatistics
tlumley at u.washington.edu        University of Washington, Seattle
^^^^^^^^^^^^^^^^^^^^^^^^
-- NOTE NEW EMAIL ADDRESS

Hola!

Thomas Lumley wrote:

On Wed, 11 Dec 2002, Michael Friendly wrote:

I'm trying to form contingincy tables among a set of character variables
which were read from a .csv file and
have missing represented as "".  I want to exclude the missing levels
from the table.

I think this is a bug. The exclude= argument doesn't work for factors,
because the argument is passed to factor(), and its exclude argument has
a different format when the main argument is a factor.

I dot think that is correct. table doesnt call factor on factors, so the
exclude argument doesnt get used at all. I had the oposite problem as
the original poster.

To include NA's as a level in a table. Why should't table simply call
factor also on factor arguments, so the exclude argument to table get
used in this case to. I modifyed table to Table doing this, and it
works, the only problem being that NA is not labelled in the table.