Printable Version of Topic

Posted by: SirAndy Jul 22 2004, 12:30 PM

ok, here's the problem:

let's say i have a table with 2 rows like the example below. in the second row, i have a integer status value (iStatus). i need a query that allows me to find the value that occurs the most in the table, in the example below, that would be "3". i'm not interested in how many 3s there are, all i need is to find out which iStatus value shows up the most in the table ...




Attached image(s)

Posted by: lapuwali Jul 22 2004, 12:36 PM

Oracle, MySQL, SQLServer, DB2, etc? It will make a difference. There is no single "SQL".

Just off the top of by non-guru head, I'd say you're looking at using MAX(COUNT(*)), probably with a sub-select, but I can't give you exact syntax w/o knowing which SQL, and I only know MySQL and Oracle (and I don't know Oracle well).

Posted by: SirAndy Jul 22 2004, 12:40 PM

Posted by: Cloudbuster Jul 22 2004, 01:18 PM

select top 1 count(iStatus) as StatCount, sDescription from {tablename} group by sDescription
order by StatCount

You'll need to replace "{tablename}" with the actual name

Posted by: SirAndy Jul 22 2004, 01:28 PM

Posted by: Deluxe247 Jul 22 2004, 01:28 PM

To get the mode do this:

SELECT TOP 1 COUNT(*) AS 'Number of repeats', rowName AS Number
FROM dbo.tblName
GROUP BY rowName
ORDER BY COUNT(*) DESC

It will return how many times the number occured and which number. You can then pull whatever you want from the result; the actual number.

Posted by: SirAndy Jul 22 2004, 01:38 PM

Posted by: vortrex Jul 22 2004, 01:41 PM

make a text file.

cat filename | cut -f2 -d" " | sort -rn | uniq | head -1

Posted by: SirAndy Jul 22 2004, 02:02 PM