SQL COUNT modify - sql

Question

From the SQL Oracle HR scheme I used the folowing:
SELECT DEPARTMENT_ID, ROUND(AVG(SALARY),2)
FROM EMPLOYEES
WHERE DEPARTMENT_ID IS NOT NULL
GROUP BY DEPARTMENT_ID
ORDER BY DEPARTMENT_ID
To get:
DEPARTMENT_ID ROUND(AVG(SALARY),2)
10 4400
20 9500
30 4150
40 6500
50 3475,56
60 5760
...
How do I change it so: it only count the departments that have the max avg salary (in my case 1) and show also the max avg salary?
Thank you for your time!

Answer 1

If I understood you, this is one possible way:
SELECT DEPARTMENT_ID, ROUND(AVG(SALARY),2) AS AVG_SALARY
FROM EMPLOYEES
WHERE DEPARTMENT_ID IS NOT NULL
AND ROUND(AVG(SALARY),2) = (
SELECT MAX(T.AVG_SALARY)
FROM (
SELECT DEPARTMENT_ID, ROUND(AVG(SALARY),2) AS AVG_SALARY
FROM EMPLOYEES
WHERE DEPARTMENT_ID IS NOT NULL
GROUP BY DEPARTMENT_ID) AS T)
GROUP BY DEPARTMENT_ID
ORDER BY DEPARTMENT_ID
This will show you ALL THE DEPARTMENTS that have max avg salary. If you want only the count:
SELECT COUNT(A.*), AVG(A.AVG_SALARY)
FROM (
SELECT DEPARTMENT_ID, ROUND(AVG(SALARY),2) AS AVG_SALARY
FROM EMPLOYEES
WHERE DEPARTMENT_ID IS NOT NULL
GROUP BY DEPARTMENT_ID) A
WHERE A.AVG_SALARY = (
SELECT MAX(T.AVG_SALARY) AS MAX_AVG_SALARY
FROM (
SELECT DEPARTMENT_ID, ROUND(AVG(SALARY),2) AS AVG_SALARY
FROM EMPLOYEES
WHERE DEPARTMENT_ID IS NOT NULL
GROUP BY DEPARTMENT_ID) AS T)

Answer 2

Another way that should work, using joins:
SELECT t1.DEPARTMENT_ID, ROUND(AVG(t1.SALARY),2) AS AVG_SALARY
FROM EMPLOYEES t1
LEFT JOIN (SELECT MAX(AVG_SALARY) AS MAX_AVG_SALARY FROM EMPLOYEES) t2
ON AVG_SALARY=t2.MAX_AVG_SALARY
WHERE t1.DEPARTMENT_ID IS NOT NULL
AND AVG_SALARY=t2.MAX_AVG_SALARY
ORDER BY t1.DEPARTMENT_ID ASC;
I tested the idea on a sample table of mine using oracle, and on a w3schools sql testing page with this code:
SELECT Customers.CustomerName, Orders.maxid
FROM Customers
LEFT JOIN (select max(Orders.CustomerID) as maxid from Orders) orders
ON Customers.CustomerID=Orders.maxid
where customers.customername is not null and customers.customerid=orders.maxid
ORDER BY orders.maxid desc;
It should grab only the departments that match their average salary with the max average salary that was selected in the join statement.
If you are only looking for the count of the departments, and not a list of the department names, then this slight modification should work for you:
SELECT COUNT(t1.DEPARTMENT_ID) as Num_Of_Depts, ROUND(AVG(t1.SALARY),2) AS AVG_SALARY
FROM EMPLOYEES t1
LEFT JOIN (SELECT MAX(AVG_SALARY) AS MAX_AVG_SALARY FROM EMPLOYEES) t2
ON AVG_SALARY=t2.MAX_AVG_SALARY
WHERE t1.DEPARTMENT_ID IS NOT NULL
AND AVG_SALARY=t2.MAX_AVG_SALARY
GROUP BY AVG_SALARY;