i am getting the error in the last proc statement

[Shrutibhatnagar]

proc sql;

      create table work.mismatches1 as

      select distinct h.product_id,churn_month

from class1.smb as h

    left  join churners as b

    on h.product_id=b.product_id; 

          quit;

proc sql outobs=200 ;

select product_id ,

case

      when churn_month is not null then '1'

      else '0'

      end as new_column

from work.mismatches1;

quit;

proc sql outobs=200;

select * ,mismatches1.new_column 

from class1.smb 

left join mismatches1 

on smb.product_id=mismatches1.product_id;

quit;

ERROR: Column new_column could not be found in the table/view identified with the correlation name MISMATCHES1.

[Kalind_Patel]

Error itself says that column new_column is not available in your mismatches1 table, check the table columns of mismatches1.

[Shrutibhatnagar]

the second proc sql contains the mismatches table which gets executed that means the table contains the new_column . where am i wrong ?

[Kurt_Bremser]

Look at the first select that creates mismatches1. The column is not included in that, so it won't be included in the dataset.

[Astounding]

The second SQL generates a report only. It doesn't create a table. There is no table that contains new_column. To create a table begin with

Create table ........

just like you did in the first sql.