topic How to count how many times value in a table repeated ? in SAS Data Management

How to count how many times value in a table repeated ?

jarr25 — Mon, 12 Jan 2015 14:34:19 GMT

Hello , I'd like to ask a question, since I'm trying to acquire result which I want, and don't know how. I have a table, and what I'd like to do, is to count how many times, value in a colums repeats. Where can I find such an option?

Let's say I have

Col1	Col2	Col3	Col4
1	1	5	g
1	2	4	b
1	3	4	b
2	4	7	d

.. and what I want to do with that is that kind of result:

there are 3 col2 that can be assigned to the value of col1 ,

there is 4 - which is number repeated most of col3 - 4 have a value of b

edit. Is the function Frequencies can count this? (how many times some value repeated?)

Re: Counting positions in a table

Jaheuk — Mon, 12 Jan 2015 14:38:57 GMT

COUNTC function ??

can you give some examples or your test data step ??

Herman

Re: How to count how many times value in a table repeated ?

data_null__ — Mon, 12 Jan 2015 14:54:24 GMT

It is difficult to understand what you want but maybe something like this.

data a;
   input col1-col3 col4 $;
   cards; 
1  1  5  g
1  2  4  b
1  3  4  b
2  4  7  d
;;;;
   run; 
proc summary descend descendtypes chartype data=a;
   class col:;
   ways 1; 
   output out=counts;
   run; 
proc print; 
   run;

Re: How to count how many times value in a table repeated ?

jarr25 — Mon, 12 Jan 2015 15:48:06 GMT

Ok I think now I will translate it better to you

I have that kind of table ^

What I need is that kind of table, but in some way made in SAS

Re: How to count how many times value in a table repeated ?

RW9 — Mon, 12 Jan 2015 16:01:33 GMT

Well, they look like group bys to me:

data have;
shiftid=3;
    employeeid=72;
    departmentid=15;
    name="Shipping and Recieving";
    output;
    employeeid=73;
    departmentid=7;
    name="Production";
    output;
run;

proc sql;
create table WANT as
select A.*,
          B.*
from    (select SHIFTID,count(distinct EMPLOYEEID) as EMPLOYEES from WORK.HAVE group by SHIFTID) A
left join (select SHIFTID,DEPARTMENTID,NAME,count(distinct EMPLOYEEID) as NUM_EMPLOY from WORK.HAVE group by SHIFTID,DEPARTMENTID,NAME) B
on      A.SHIFTID=B.SHIFTID;
quit;

Re: How to count how many times value in a table repeated ?

data_null__ — Mon, 12 Jan 2015 16:07:12 GMT

Show data in data step with CARDS. Not a picture.

Re: How to count how many times value in a table repeated ?

Reeza — Mon, 12 Jan 2015 16:51:22 GMT

You'll need to modify the variable and table names but the following is

proc sort data=table1;

by shiftID employee department;

run;

proc freq data=table1;

by shiftID;

table department;

run;

Re: How to count how many times value in a table repeated ?

Astounding — Mon, 12 Jan 2015 17:23:01 GMT

If you want something quick and dirty, you can use:

proc freq data=table1;

tables ShiftID * DepartmentID * Name / missing list out=counts;

run;

If you want something pretty, you'll have to take the output data set COUNTS and figure out how to print it. It will already contain the data that you need, and it's just a matter of generating the proper report.

Good luck.