0
$\begingroup$

I have two tables (matrix) of same dimensions, one contains correlation coefficients and other with p values. I want to combine them into one table. For example let's say I have correlation coefficient between variable A1 and A2 of 0.75 in table 1 and p value of 0.045 in table 2. Now in my combined table 3, I want to use:

condition1 for table 1: if a coefficient value in a cell of table 1 is less than 0.4 then "+", 0.4 <= coefficient <0.7 then "++" else "+++",

condition2 for table 2: if a pvalue in a cell of table 2 is less than 0.01 then "+++", 0.01 <= pvalue < .05 then "++" else "+".

Thus corresponding cell value for A1 and A2 in table 3 should look like: +++/++ where "+++" correspond to table 1 value of 0.75 and ++ correspond to table 2 p value of 0.045 and "/" is just a separator.

I would like to do this either is SAS or R. Thanks in advance.

$\endgroup$
1
  • $\begingroup$ There is not much statistics in your question, so perhaps better suited for Stack Overflow... but what the hell, I'll answer anyway. $\endgroup$
    – Nick Sabbe
    Sep 22 '11 at 15:06
1
$\begingroup$

I think this should come close in R (untested since you did not provide data):

t1val<-ifelse(T1<0.4, "+", ifelse(T1<0.7, "++", "+++"))
t2val<-ifelse(T2<0.01, "+", ifelse(T2<0.05, "++", "+++"))

t3<-paste(t1val, t2val, sep="/")
t3<-array(t3, dim=dim(t1))

Basically, you treat the original matrices as vectors and then at the end convert the result back.

$\endgroup$
2
  • $\begingroup$ thanks, so t3<-paste(t1val, t2val, sep-"/") right? $\endgroup$
    – absha
    Sep 22 '11 at 15:18
  • 2
    $\begingroup$ Check out the symnum function - it can save a lot of ifelses. $\endgroup$
    – Aniko
    Sep 22 '11 at 16:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.