Count Columns in a row and Logic Relates

Question

I need some help in oracle SQL,
Student(unique) has five races in a table
I need a select statement with student_id - 1 column
I need to count the number columns with values 'Yes' in them
the columns are race1,race2,race3,race4,race5 for a single row --- this should be 2 ndcolumn
The 3rd column should contain the max of the race having 'yes' in them
i.e if race1 and race 3 have values'yes' in them i need 3 in the next column, - 3rd column
Can some one help me in this

Answer 1

Hi,
As pointed out by John, the datamodel is wrong and keeping it as separate table with (student_id, race_id) will be ideal. But, if you still want to use your existing data model, you can use something like
WITH students AS (SELECT 1 student_id,
                         'Yes' race1,
                         'No' race2,
                         'Yes' race3,
                         'No' race4,
                         'Yes' race5
                    FROM DUAL
                  UNION ALL
                  SELECT 2 student_id,
                         'Yes' race1,
                         'No' race2,
                         'Yes' race3,
                         'Yes' race4,
                         'No' race5
                    FROM DUAL)
SELECT student_id,
       DECODE (race1, 'Yes', 1, 0)
    + DECODE (race2, 'Yes', 1, 0)
    + DECODE (race3, 'Yes', 1, 0)
    + DECODE (race4, 'Yes', 1, 0)
    + DECODE (race5, 'Yes', 1, 0) count_race,
       CASE
         WHEN race5 = 'Yes' THEN 5
         WHEN race4 = 'Yes' THEN 4
         WHEN race3 = 'Yes' THEN 3
         WHEN race2 = 'Yes' THEN 2
         WHEN race1 = 'Yes' THEN 1
         ELSE 0
       END max_race
  FROM studentsThanks,
sukhijank