how to count rows which match various words, broken  down by month

select month(txttime), year(txttime), count(*) from txts
where (message like '%bully%' or message like '%bullied%' or message like '%blly%')
group by month(txttime), year(txttime)

Thanks, but this captures other words which have these letters in them too, I guess most of the time it would have spaces around words but sometimes it could be at the beginning or end of the data or also it could have a ',' or a '.' I guess, the data is TXT messages that you would send with your cell phone.. I have a list of about 100 different sets of these key words I am running in a loop...

yes, can you please tell me how to get those words as the result, I don't think using like '%word%' is good enough, how can it know if there are full stops or at the beginning etc? would using a regex be better? though this is going over 350000 rows so I guess regex would be slow, I am happy to test it though but don't know the syntax

at the beginning and end you'd have to do:

left(message, 5) = 'bully' or left(message, 4) = 'blly'
or right(message, 5) = 'bully' or right(message, 4) = 'blly'

etc

if you want to check for spaces, you'd have to do message like '% bully %'

using regex (assuming you're using mysql): http://www.tech-recipes.com/rx/484/use-regular-expressions-in-mysql-select-statements/

Thanks, could be explained with better examples though but I can figure the rest now out cheers :)