topic Re: Creating a new unique ID variable from two variables in merged data with repetition of one varia in SAS Programming

Creating a new unique ID variable from two variables in merged data with repetition of one variable

gonzy31 — Thu, 03 Aug 2017 02:04:12 GMT

Hi all,

Would love some help with this, as I think I am doing something wrong here in SAS. I merged multiple years of data, but I wanted to create a unique ID variable ("ID") to identify each unique physician ("phycode"). For each individual year, the "phycode" is unique for every physician. However, the "phycode" is repeated in subsequent years for other physicians so both "phycode" and "year" are necessary to create a new unique ID in the dataset.

Here is what my data looks like now:

Phycode Year

1 2004

1 2005

2 2004

2 2005

3 2004

4 2004

I want my output to produce a table like this in recognition that repeat phycodes in different years represent unique physicians (ID):

Phycode Year ID

1 2004 1

1 2005 2

2 2004 3

2 2005 4

3 2004 5

4 2004 6

Here is what I tried:

proc sort data=IN.NAMCS0514;

by phycode year;

run;

data=UniqueID;

set NSS0514;

by phycode year;

if first.phycode then ID=1;

ID +1;

run;

Thanks in advance for your help

Re: Creating a new unique ID variable from two variables in merged data with repetition of one varia

Astounding — Thu, 03 Aug 2017 03:20:18 GMT

Well you could replace the last two statements with:

if first.year then ID + 1;

Alternatively, you don't really need an ID variable, You could just use the combination of PHYCODE and YEAR to identify a physician.

Re: Creating a new unique ID variable from two variables in merged data with repetition of one varia

ChrisNZ — Thu, 03 Aug 2017 03:46:13 GMT

Your output matches:

data UniqueID;

set NSS0514;

ID +1;

run;

Re: Creating a new unique ID variable from two variables in merged data with repetition of one varia

gonzy31 — Thu, 03 Aug 2017 12:42:38 GMT

Thank you both! That worked perfectly.