topic Re: double observations in SAS Programming

double observations

vraj1 — Fri, 29 Sep 2017 09:29:34 GMT

I have the below code with conditions and when i run it it creates double observations.

I am creating a dataset which gives 1 if it satisfies a condition for each treatment group. But it resolves seperately and creates duplictes.

i use the below condition to have treatment wise but it duplicates like below in my datastep

%** Loop for each arm (to avoid transposing the data later) **;
%do k = 1 %to %sysfunc(countw(&ArmList., #)); %** Start for each ArmVar **;
if compress(&ArmVar.) = "%scan(&ArmList.,&k., #)" then do; %** if ArmVar condition **;

and it is resolved to code attached.

Can any one help me what i am doing wrong

Re: double observations

FredrikE — Fri, 29 Sep 2017 10:32:13 GMT

Check subjid S1287, it will be outputted 3 times, the reason is that it fulfills these conditions:

if compress(actarm) = "BLINDED"

if SAFFL='Y'

if dsreas_num_BLINDED

All of these has an output statement, you have to check your conditions again if this is not the desired result 🙂

//Fredrik

Re: double observations

vraj1 — Fri, 29 Sep 2017 10:33:46 GMT

If it fulfills all conditions then it should be multiple. But what i get is multiple obs of subjids for almost most of them

Re: double observations

FredrikE — Fri, 29 Sep 2017 10:49:42 GMT

I think you need to pick some of the obs you don't want to have multiple times and check the rules for them 🙂

Re: double observations

Kurt_Bremser — Fri, 29 Sep 2017 11:09:41 GMT

@vraj1 wrote:

If it fulfills all conditions then it should be multiple. But what i get is multiple obs of subjids for almost most of them

Then they all fulfill multiple conditions. Maxim 3: Know your data.