Put row info to columns with yes and no

[jonatan_velarde]

Good day SAS friends:

i have a data set containing information from individuals from school, and i know what they eat at lunch, each information is separated by semicolon ";", as follows

data have;

input student answers$;

cards;

1 bread; soda

2 bread; soda; water

3 water; bread

4 soda; crackers

5 crackers

6 crakers; water

;

student	bread	soda	water	crackers
1	yes	yes	no	no
2	yes	yes	yes	no
3	yes	no	yes	no
4	no	yes	no	yes
5	no	no	no	yes
6	no	no	yes	yes

Thanks in advance

the new data set that i want to create based in this is:

[Kurt_Bremser · Accepted solution]

First, create a long dataset:

data long;
set have;
answer = "yes";
do i = 1 zo countw(answers,';');
  meal = scan(answers,i,';');
  output;
end;
keep student meal answer;
run;

Then transpose to wide:

proc transpose data=long out=wide;
by student;
var answer;
id meal;
run;

If necessary, you can replace any missing values in a follow-up step with "no". The beauty of the transpose is that you do not need to know all the meals beforehand.

[novinosrin]

Hi @jonatan_velarde  By all means @Kurt_Bremser  's solution is indeed clean as it is data driven that gives you the comfort of not having to be concerned about what's in your data. However, should you know beforehand that there exists  only 4 distinct lunch items in your data, i.e.   

bread

soda

water

crackers

you may consider going lazy declaring an array like-

data have;
input student answers & $20.;
cards4;
1 bread; soda
2 bread; soda; water
3 water; bread
4 soda; crackers
5 crackers
6 crakers; water
;;;;

data want;
 set have;
 array food(*)$10 bread	soda water crackers;
 array t(4) $10 _temporary_ ('bread','soda','water','crackers');
 do _n_=1 to dim(food);
  food(_n_)=ifc(find(answers,strip(t(_n_))),'YES','NO');
 end;
 drop answers;
run;