topic Re: Summing values of a variable in proc tabulate in SAS Programming

Summing values of a variable in proc tabulate

div44 — Tue, 06 Feb 2018 20:51:28 GMT

Hello,

I have a variable x which is a binary variable (may have missing values). A sum function in proc means would give me value of 60. (yes=60, no or missing=40).

When I use the sum function in proc tabulate with a class statement, I get the number of obs (n=100) instead of the sum (60).

Here is the code I am using:

proc tabulate data = have;

var x;

class y;

table x*sum, y;

run;

Any help would be appreciated.

Thanks

Re: Summing values of a variable in proc tabulate

ballardw — Tue, 06 Feb 2018 21:19:32 GMT

Show your output or provide the data used in the form of a data step, preferably both.

Instructions here: https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-data-AKA-generate/ta-p/258712 will show how to turn an existing SAS data set into data step code that can be pasted into a forum code box using the {i} icon or attached as text to show exactly what you have and that we can test code against.

Re: Summing values of a variable in proc tabulate

Reeza — Tue, 06 Feb 2018 22:32:58 GMT

Your code works fine for me. You have some other issue.

If you run the following the results should be the following. I added in the Count/N so you can see the difference and the PROC MEANS to show that the results are correct.


		y
		0	1
x	Sum	20.00	12.00
x	N	56	44

data have;
    call streaminit(12345);

    do i=1 to 100;
        x=rand('bernoulli', 0.4);
        y=rand('bernoulli', 0.5);
        output;
    end;
run;

proc tabulate data=have;
    var x;
    class y;
    table x*sum x*n, y;
run;

proc means data=have n sum nway;
    var x;
    class y;
run;

@div44 wrote:

Hello,

I have a variable x which is a binary variable (may have missing values). A sum function in proc means would give me value of 60. (yes=60, no or missing=40).

When I use the sum function in proc tabulate with a class statement, I get the number of obs (n=100) instead of the sum (60).

Here is the code I am using:

proc tabulate data = have;

var x;

class y;

table x*sum, y;

run;

Any help would be appreciated.

Thanks

Re: Summing values of a variable in proc tabulate

sasuser123123 — Wed, 24 Jul 2019 09:22:29 GMT

How to get sum of a variable..by group.it means I've Male and female in one variable like 'sex' and values in another variable like 'aval' so I want the result like

Gender
--------------
F. M
---------------
Aval. Aval
-------------------
Sum. Sum
--------------------

Re: Summing values of a variable in proc tabulate

sasuser123123 — Wed, 24 Jul 2019 09:23:04 GMT

Sorry I've mentioned Gender instead of sex

Re: Summing values of a variable in proc tabulate

Patrick — Wed, 24 Jul 2019 10:16:13 GMT

As @Reeza demonstrated your code should work. If this is not the case then you need to post a SAS data step which creates sample data, your proc which creates the report with the "issue" using your sample data - and then explain us what result you want instead.

Re: Summing values of a variable in proc tabulate

sasuser123123 — Wed, 24 Jul 2019 12:47:25 GMT

data is

sub sex site trt avisit paramcd aval

001 F USA palcebo week0 alt 9

002 F USA sudy_dr week0 ast 10

003 M USA palcebo week4 alt 11

004 F USA palcebo week4 ast 10

005 M ex-USA sudy_dr week0 alt 16

006 M USA palcebo week0 ast 11

007 F ex-USA sudy_dr week4 alt 13

008 M USA palcebo week0 ast 16

009 F ex-USA sudy_dr week0 alt 10

010 M USA palcebo week4 ast 17

I need Output like

Avisit

week0 week4

sex all sex all

F M F M

site

USA TRT

placebo 50 . 50 50 . 50

study_dr . 50 50 . 50 50

all 100 100 100 100 100 100

ex-USA TRT

placebo 25 50 75 25 50 75

study_dr 25 25 25 25

all 100 100 100 100 100 100

please find out the solution....using proc tabulate .....actually I've 24 observations but I took only 10..

Re: Summing values of a variable in proc tabulate

Reeza — Wed, 24 Jul 2019 15:00:59 GMT

You should start your own question, not on a thread that's a year old.