Hi
This is a mixed approach. I've successfully used Tony's Column Specific Filtering technique for other aspects of my survey.

In this case, I have a number of subquestions that are rated (Array type). When a user rates a subquestion, we use an array filter to then pass the subquestions down into a Ranking Type question.

The issue is that one of the ratings is 'Useless' and so ideally we'd like to NOT have an item rated as 'Useless' put into the question that ranks them based on their value (given the respondent has already done that). The array filter isn't that flexible.

Is there a routine that could help pull these items out of the rank table?

(BTW: there is already a fair bit of Java script on these questions (plus a relevance equation on each question which I've removed here so they show), and I have three similar to the one attached on a page so it is taking a while to load . Any ideas to speed up, or to flash a 'please wait' would be appreciated!)

Thanks for pondering, Lloyd.

Just an update. I discussed this problem with TPartner offline, and he pointed out the challenges of what I'm asking.

Having filtered the data, the problem would arise where a person subsequently goes back to the rating table and changes an item to 'useless.' This would require the code to shift a rated item back to the left column(unrated) and then delete/hide it. This could also take that ranking question below the 5 minimum responses required for its validation....
so perhaps not a realistic proposition (unless you prevent someone going back!).

May pay to rewrite the questions.

And the slow load seems to have resolved this week - perhaps just too many questions on the page so breaking it up helped.

Lloyd.