I always try to use SUPRTOOL (if you have it) to extractand sort all subsets of
 data rather than QUIZ links and then just use Quiz for the final report.  The
 data selection is usually clearer than multiple complex QUIZ links, and the
 data extraction is always faster.

for example:
!QUIZ
;------------------------------------------------------------
;Create an almost empty Quiz subfile (ARCBOARD)
;------------------------------------------------------------
ACCESS ARC-DETL
SET REPORT LIMIT 1
REPORT SUMMARY &
ARC-SSN ARC-GRADE ARC-COURSE ARC-LASTTRAN
SET SUBFILE NAME ARCBOARD SIZE 1000000
GO
EXIT
!RUN SUPRTOOL.PUB.ROBELLE
{------------------------------------------------}
{Fill the ARCBOARD subfile with     }
{data from ARC-DETL                        }
{------------------------------------------------}
BA DBASE,5,READPASS
GET ARC-DETL
IF (ARC-COMPONENT="2" AND( ARC-GRADE>="E5" AND ARC-GRADE<="E8") AND
 ARC-COURSE[3],4="00" AND ARC-LASTTRAN[1],1 ="Z")
EXTRACT &  ARC-SSN,ARC-GRADE,ARC-COURSE,ARC-LASTTRAN
PURGE ARCBOARD
OUT ARCBOARD,ERASE
XEQ
{------------------------------------------------}
{Append to the ARCBOARD subfile}
{with data from STUD-CRS-DETL-H}
{------------------------------------------------}
GET STUD-CRS-DETL-H
 IF (COMPONENT="2" AND (GRADE>="E5" AND & GRADE<="E8") AND COURSE[3],4="00" AND
 & LASTTRAN[1],1 ="Z")
EXT SSN, GRADE,COURSE,LASTTRAN
OUT ARCBOARD,APPEND
XEQ
IN ARCBOARD
DEFINE SUPR-SSN,1,9
SORT SUPR-SSN
O = I
EXIT
!QUIZ
ACCESS *ARCBOARD LINK TO STUD-INFO-DETL
... Quiz statements to do final reporting

This makes several assumptions, but the general methodology should work.

p.s. You may also want to surround your SELECT statement with parenthesis where
 apropriate.

Charlie Fontaine      NYLCare Health Plans