Counting values in multiple variables to create new variable

[JenMMerc]

Hi,

I am working with a data set where I have created new variables based on success of outcomes. 0=not sucessful 1=successful .= outcome not applicable to client.  I want to create a 2 new variables count_1 and count_0 that each count the number of 1s and 0s from the outcomes variables so I can eventually calculate a success rate for each client. Thanks in advance for the help.

data outcome;

input client outcome1 outcome2 outcome3;

cards;

A . . 0

B . 1 0

C 0 0 1 

D . . 1

;

run;

[ballardw · Accepted solution]

In a datastep

data want;

   set have;

   count_1= sum( of outcome: );

   count_0 = n(of outcome: ) - count_1;

run;

the outcome: with the colon is a variable list of all of the variables, with the data step functions such as sum, mean, n, std etc then

sum ( of outcome:) is the sum of all the non-missing values. Which with 1/0 coded variables is the same as the count of 1.

The N function returns the number of non-missing values.

 BTW mean(of outcome: ) will the be percent of 1's .

[novinosrin]

Very neat and elegant sir @ballardw 

FWIW an ordinary solution

 
data want;
set outcome;
count_1=countc(cats(of outcome:),'1');
count_0=countc(cats(of outcome:),'0');
run;

[JenMMerc]

It worked! Thank you.

[novinosrin]

Hello @JenMMerc 

Welcome to SAS communities. May i request you to mark BallardW's solution as answered and close the thread. Thank you!