Manipulating Duplicate Records

I have a need to identify duplicate Employee records in a temp file, based on USERID. Then I need to look at the duplicate record sets and look at the Emp Status:
1. If both (or all) records are INACTIVE, I do not need to report on them (i.e ignore them).

2, If a set (of duplicate records) has just one ACTIVE record, ignore the whole set.

3. If both (or all) records are ACTIVE, then report on them.

Example set:

{userid}              {Employee}                  {Status}                      {Action}

{ABCD}                {1234}                         {INACTIVE}
{ABCD}                {4567)                         {INACTIVE}                  {Ignore the pair}

{EFGH}                 {9876}                         {ACTIVE}
{EFGH}                 {5432}                         {INACTIVE}                 {Ignore the pair}

{IJKL}                   {8315}                          {ACTIVE}
{IJKL)                   (5138}                          {ACTIVE}                     {Report on them}

I will appreciate any suggestions as to how to code this in SQL (Oracle).

Thanks.
Maalim