topic Re: Inserting Columns into SAS Datasets in SAS Programming

Inserting Columns into SAS Datasets

maroulator — Fri, 05 Sep 2014 16:16:24 GMT

Does anyone know if there is a way (ie SAS command) to insert a new variable between existing SAS variables in a dataset?

For example I have the following dataset......

Name Age

Tom 10

John 18

Jill 23

Chris 25

Don 31

......and I want to end up with the dataset below.

Name Sex Age

Tom M 10

John M 18

Jill F 23

Chris M 25

Don M 31

Re: Inserting Columns into SAS Datasets

Jagadishkatam — Fri, 05 Sep 2014 16:19:30 GMT

if you have another dataset with name and sex variables , then you can use the merge statement to combine the age dataset with gender dataset by name variable and you can get the three variables in a single dataset.

something like

data want;

merge age gender;

by name;

run;

Thanks,

Jag

Re: Inserting Columns into SAS Datasets

stat_sas — Fri, 05 Sep 2014 16:21:46 GMT

Try retain statement after adding variable to your existing dataset.

data want;

retain Name Sex Age;

set have;

run;

Re: Inserting Columns into SAS Datasets

Hima — Fri, 05 Sep 2014 16:35:35 GMT

What ever Jagadishkatam and stat@sas provided works but for my understanding is the variable Sex on another table. If so, what is primary key and foriegn key relationship between those tables. Can you elobarate?

Re: Inserting Columns into SAS Datasets

maroulator — Fri, 05 Sep 2014 16:57:02 GMT

Hima,

There is no relationship per se. I have a dataset that already contains the Name and Age variables and I am trying to create/generate a new variable (Sex) between the two existing ones. I was hoping that there is a way to do this in SAS, similar to how you can insert a new column in Excel and fill it in with values.

I know how to create a new variable at the end of the variables in my existing dataset, but I was hoping to be able to create it betweent the two existing ones.

Re: Inserting Columns into SAS Datasets

Hima — Fri, 05 Sep 2014 17:19:34 GMT

Thank you so much. I created a scenario to myself as per your explanation. Please adjust it your needs. What stat@sas provided works great in the situation.

DATA HAVE;
INPUT NAME $    AGE;
DATALINES;
TOM         10
JOHN        18
JILL        23
CHRIS       25
DON         31
;
RUN;

DATA WANT;
RETAIN NAME SEX AGE;
SET HAVE;

IF INDEX(NAME,'O') THEN
SEX='M';
ELSE SEX='F';
RUN;