What is the best way to fix this SQL? What I did is to ADD another where clause below: and (((COUNT(m.PROC_CD) / 15780) * 100 - AVG_CLAIM_PCT)/STDDEV_CLAIM_PCT) > 2 into this SQL and keep getting group function is not allowed here. Thanks, Vu select RTRIM(PROV_ID), RTRIM(STDDEV_CLAIM_PCT), RTRIM(STDDEV_DOL_PCT), RTRIM(PROC_GRP), RTRIM(m.proc_cd) as PROC_CD, LTRIM(TO_CHAR(COUNT(m.PROC_CD), '999,999,999,999')) as NUM_SRV, LTRIM(TO_CHAR(SUM(PAID_AMT), '$999,999,999,999.99')) as DOLLARS, LTRIM(TO_CHAR((COUNT(m.PROC_CD) / 15780) * 100, '999,999,999,999.9')) as PCT_SRV, LTRIM(TO_CHAR((SUM(PAID_AMT) / 437780.94000000006) * 100, '999,999,999,999.9')) as PCT_DOL, LTRIM(TO_CHAR(avg(PEER_CT), '999,999,999,999')) as PROV_CT, LTRIM(TO_CHAR(AVG_CLAIM_PCT, '999,999,999,999.99')), LTRIM(TO_CHAR(((COUNT(m.PROC_CD) / 15780) * 100 - AVG_CLAIM_PCT)/STDDEV_CLAIM_PCT, '999,999,999,999.99')) as Z_CLAIM_PCT, LTRIM(TO_CHAR(AVG_DOL_PCT, '999,999,999,999.99')), LTRIM(TO_CHAR(((SUM(PAID_AMT) / 437780.94000000006) * 100 - AVG_DOL_PCT)/STDDEV_DOL_PCT, '999,999,999,999.99')) as Z_DOL_PCT from spy_mart m , spy_proc_norm n where SET_ID = '110' AND PROV_ID = '111909101' and m.proc_cd=n.proc_cd and n.model_id = 'GENRIC04' and n.peer_id='30' and (((COUNT(m.PROC_CD) / 15780) * 100 - AVG_CLAIM_PCT)/STDDEV_CLAIM_PCT) > 2 group by m.proc_cd, PROC_GRP, PROV_ID, AVG_CLAIM_PCT, AVG_DOL_PCT, STDDEV_CLAIM_PCT, STDDEV_DOL_PCT order by NVL(((COUNT(m.PROC_CD) / 15780) * 100 - AVG_CLAIM_PCT)/STDDEV_CLAIM_PCT, -999999999) desc