Consider the following SQL Queries executed on the relation Employee:

Employee (Eid, Ename, sal)

Eid is the primary key


S1:SELECT count (*) FROM Employee;

S2: SELECT count (Eid) FROM Employee;

Will these queries return the same result or different results?

According to me, these queries should return same values.

But my solution manual claims that the values returned should be different. It says that S1 will also include results where Eid field is empty.

But according to me, the primary key field can never be empty. So, there will never be a case where Eid field will be empty. As this field will never be empty, the output of S1 and S2 should be equal.

Can anyone please tell me whether I am correct? If not, please tell me what I am doing wrong.

Thanks in advance!

  • 2
    $\begingroup$ Obviously, it totally depends on your Eid whether it is unique and not null. If it is primary key the result will be same. $\endgroup$ Commented Dec 28, 2018 at 19:09

1 Answer 1


where Eid field is empty.

What do you mean by "empty"? Perhaps you mean 'null'?

What words exactly does your "solution manual" use? "empty" or "null"?

If the Eid field is CHAR there could easily be a row with Eid "empty" -- that is: zero length, or containing only space characters. count(*) and count(Eid) return the same value in that case.

If the Eid field is NULLABLE there could easily be a row with Eid null. count(*) will include that row, whereas count(Eid) will not.

Usually fields in keys are expected to be NOT NULL. Some vendors' implementations of the SQL standard insist on NOT NULL. But some don't. It is good practice to make key fields NOT NULL. Presumably your manual is getting you to think about why.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.