SQL - counting max number of employees and excluding string using subqueries - sql

Question

These are my two tables:
employee
employee_id employee_name job manager_id hire_date salary commission department_id
----------------------------------------------------------------------- -------------------------------------
7839 KING PRESIDENT 20-NOV- 01 5000 50
7596 JOST VICE PRESIDENT 7839 04-MAY- 01 4500 50
7603 CLARK VICE PRESIDENT 7839 12-JUN- 01 4000 50
7566 JONES PUBLIC ACCOUNTANT 7596 05-APR- 01 3000 10
7886 STEEL PUBLIC ACCOUNTANT 7566 08-MAR- 03 2500 10
7610 WILSON ANALYST 7596 03-DEC- 01 3000 20
7999 WOLFE ANALYST 7610 15-FEB- 02 2500 20
7944 LEE ANALYST 7610 04-SEP- 06 2400 20
7900 FISHER SALESMAN 7603 06-DEC- 01 3000 500 30
7921 JACKSON SALESMAN 7900 25-FEB- 05 2500 400 30
7952 LANCASTER SALESMAN 7900 06-DEC- 06 2000 150 30
7910 SMITH DATABASE ADMINISTRATOR 7596 20-DEC- 01 2900 40
7788 SCOTT PROGRAMMER 7910 15-JAN- 03 2500 40
7876 ADAMS PROGRAMMER 7910 15-JAN- 03 2000 40
7934 MILLER PROGRAMMER 7876 25-JAN- 02 1000 40
8000 BREWSTER TBA 22-AUG- 13 2500
department
department_id department_name address
------------- -------------------- --------------------
10 ACCOUNTING NEW YORK
20 RESEARCH DALLAS
30 SALES CHICAGO
40 IT DALLAS
50 EXECUTIVE NEW YORK
60 MARKETING CHICAGO
I need to write a statement using subqueries (not joins) to display the name and address of all departments (excluding departments in Dallas) having the maximum number of employees. I cannot have any hard coding except the string 'DALLAS'
The result should look like this:
DEPARTMENT_NAME ADDRESS
-------------------- --------------------
EXECUTIVE NEW YORK
SALES CHICAGO
I'm at a loss on doing this without joins. Here are my three tries:
SELECT department_name, address
FROM department
WHERE department_id NOT IN
(SELECT department_id
FROM department
WHERE UPPER(address) + 'DALLAS')
ORDER BY department_name;
SELECT department_name, address
FROM department
WHERE department_id NOT IN
(SELECT department_id
FROM department
WHERE UPPER(address) = 'DALLAS')
AND department_id > ALL
(select COUNT(department_id) FROM employee GROUP BY department_id)
ORDER BY department_id;
select *
from department
where department_id in (select department_id
from employee
group by department_id
having count(*) = (select max(num)
from (select department_id,
count(*) as num
from employee_tbl
where address != 'DALLAS'
group by department_id)));
Can anyone PLEASE tell me if I'm even getting close and guide me in the right direction? Thanks

Answer 1

The departments with the maximum number of employees (using a single table scan):
SELECT department_id
FROM (
SELECT department_id,
RANK() OVER ( ORDER BY num_employees DESC ) AS rnk
FROM (
SELECT department_id,
COUNT(1) AS num_employees
FROM employees
GROUP BY department_id
)
)
WHERE rnk = 1;
Finding the departments not in DALLAS:
SELECT department_id
FROM departments
WHERE address <> 'DALLAS'
So combining the two:
SELECT department_id
FROM (
SELECT department_id,
RANK() OVER ( ORDER BY num_employees DESC ) AS rnk
FROM (
SELECT department_id,
COUNT(1) AS num_employees
FROM employees
WHERE department_id IN (
SELECT department_id
FROM departments
WHERE address <> 'DALLAS'
)
GROUP BY department_id
)
)
WHERE rnk = 1;

Answer 2

maybe something like this?
select department_name, address
from department
where department_id in
(SELECT e.department_id
FROM employee e
group by e.department_id
having count(*) = (
select max(count(e2.employee_id))
FROM employee e2
WHERE department_id NOT IN (SELECT department_id
FROM department
WHERE UPPER(address) = 'DALLAS')
group by e2.department_id
)
)
and UPPER(address) != 'DALLAS'
but it is not the best query in plane :) very strange
In your query Ellen I did not understand where from the table
employee_tbl