Skip to content

contingency table

4 messages · reena, Marc Schwartz, Rolf Turner

reena
# 
Hello,

I want to create contingency table. The data which should be presented in
contingency table look like this.(Its a very long list. I am copying small
part of it)

obs1    obs2    exp1     exp2

3           8          725      875
0           0           58        70
3           7          435      525
10         7          754      910
0           1          145     175

the table should be like this.
----------|-----------|
obs1    | obs2    |
----------|-----------|
exp1    |  exp2   |
-----------------------


I have tried to use ftable function. But it didn't work. Please help me.

Regards
Reena 

--
View this message in context: http://r.789695.n4.nabble.com/contingency-table-tp4217741p4217741.html
Sent from the R help mailing list archive at Nabble.com.
reena
# 
This is my list.
obs1    obs2    exp1     exp2

3           8          725      875
0           0           58        70
3           7          435      525
10         7          754      910
0           1          145     175

and i want result in contingency table as

obs       3                8
exp    725          875

next table will be

obs    0         0
exp   58       70

and so on... 

--
View this message in context: http://r.789695.n4.nabble.com/contingency-table-tp4217741p4218845.html
Sent from the R help mailing list archive at Nabble.com.
Marc Schwartz
#
On Dec 20, 2011, at 12:36 PM, reena wrote:

            

      
      
      
    

        
Hi,

It's not clear what you intend to do with the data after restructuring, but creating a list of the 2x2 tables from each row seems a reasonable first step.

Thus, if your data above is contained in a data frame 'DF':

            

      
      
      
    

        
obs1 obs2 exp1 exp2
1    3    8  725  875
2    0    0   58   70
3    3    7  435  525
4   10    7  754  910
5    0    1  145  175


# Iterate over each row in DF, returning a 2x2 matrix from each

            

      
      
      
    

        
[[1]]
     [,1] [,2]
[1,] 3    8   
[2,] 725  875 

[[2]]
     [,1] [,2]
[1,] 0    0   
[2,] 58   70  

[[3]]
     [,1] [,2]
[1,] 3    7   
[2,] 435  525 

[[4]]
     [,1] [,2]
[1,] 10   7   
[2,] 754  910 

[[5]]
     [,1] [,2]
[1,] 0    1   
[2,] 145  175 


This gives you a list of 5 2x2 matrices, one from each row in DF. See ?lapply, ?seq and ?matrix.

HTH,

Marc Schwartz

  
  


  


      

    

    

      
      

        


  

        

      Rolf Turner
    

    

      
      #
    

  


  

      
Do you actually know what "contingency table" means?

The tables in your example make no sense at all as contingency
tables *especially if "obs" means "observed" and "exp" means
"expected".

You can, however, extra the tables in the manner which you seem
to desire, as follows:  Let your data object be called "dat".  (And
surely this object is either a data frame or a matrix --- judging by
the display --- and *NOT* a list.  Learn to use correct terminology
and you will be much more likely to get useful replies to your questions.)

dat <- as.matrix(dat) # To make sure it's a matrix r.t. a data frame.
xxx <- lapply(1:nrow(dat), 
function(i,x){matrix(x[i,],ncol=2,byrow=TRUE)},x=dat)

Now "xxx" is a (genuine!) list, and the i-th entry of this list is the 
i-th table from
your "required" set of tables.  E.g.:

xxx[[3]]
         [,1]   [,2]
[1,]      3      7
[2,]  435  525

     cheers,

         Rolf Turner

        
On 21/12/11 07:36, reena wrote: