I have the following :

CREATE TABLE [dbo].[a_table1](
    [id] [int] IDENTITY(1,1) NOT NULL,
    [caseID] [int] NOT NULL,
    [officeID] [int] NOT NULL,
    [location] [varchar](50) NULL
) ON [PRIMARY]
GO


INSERT INTO a_table1 VALUES ( 23, 44, 'hq' ) 
INSERT INTO a_table1 VALUES ( 86, 44, 'hq' ) 
INSERT INTO a_table1 VALUES ( 26, 44, 'hq' ) 
INSERT INTO a_table1 VALUES ( 32, 44, 'hq' ) 

INSERT INTO a_table1 VALUES ( 15, 34, 'satellite' ) 
INSERT INTO a_table1 VALUES ( 32, 34, 'satellite' ) 
INSERT INTO a_table1 VALUES ( 33, 34, 'satellite' ) 
INSERT INTO a_table1 VALUES ( 86, 34, 'satellite' ) 
INSERT INTO a_table1 VALUES ( 17, 34, 'satellite' ) 

SELECT * FROM a_table1 

This is the desired output :

basically, get all caseIds from hq location, and get only caseIds that do not exist on the satellite locations.

any ideas?