Extracting the first non-zero digit from a numeric variable + delete duplicates

[larkjr18]

Here is my code and I attached the output. Some of the data only extracted 3 digits and then randomly switches to extracting 4 digits?

This is the problem i am solving 

1. Consider only the first 4 character digits of SSN, find the duplicate groups through By-group processing, and create a SAS data set for these observations. Present the screenshot of the partial SAS LOG, showing the number of observations in this subset of data. Show the output for “corrected SSN” variable using PROC PRINT (OBS=8);”.

data hi.question4;
infile 'C:\Users\larkj\Desktop\STUDENT.DAT';
input ssn 1-9;
SSNc = put(ssn,9.);
drop SSN;
rename SSNc = SSN;
SSN_New=substr(SSNc,1,4);
run;
proc sort data = hi.question4
dupout = SSN_New
nodupkey;
by SSN_NEW;
run;
proc print data=hi.question4;
run;

[Reeza · Accepted solution]

Your datastep doesn't make sense. Read SSN as a character from start, don't convert it. 

And the subject line doesn't make sense for your question either...

[larkjr18]

That worked great, don't know why I didn't do that before. Thank you

[Reeza]

If you had wanted to keep using the previous code, use Z9 to apply a 9 digit format with leading 0s. 

[Tom]

You can read the SSN as a character string. 

input SSNc $9. ;

You can convert a numeric value to 9 digits with leading zeros using the Z9 format.

SSNc = put(ssn,Z9.);

You can get the first 4 digits of a 9 digit number using arithmetic.

prefix = int(ssn/10000);

You can get the first 4 characters of a 9 character string using SUBSTR().

prefixc = substr(SSNc,1,4);

[Reeza]

---

@larkjr18 wrote:

Here is my code and I attached the output. Some of the data only extracted 3 digits and then randomly switches to extracting 4 digits?

This is the problem i am solving 

1. Consider only the first 4 character digits of SSN, find the duplicate groups through By-group processing, and create a SAS data set for these observations. Present the screenshot of the partial SAS LOG, showing the number of observations in this subset of data. Show the output for “corrected SSN” variable using PROC PRINT (OBS=8);”.

data hi.question4;
infile 'C:\Users\larkj\Desktop\STUDENT.DAT';
input ssn 1-9;
SSNc = put(ssn,9.);
drop SSN;
rename SSNc = SSN;
SSN_New=substr(SSNc,1,4);
run;
proc sort data = hi.question4
dupout = SSN_New
nodupkey;
by SSN_NEW;
run;
proc print data=hi.question4;
run;

---