4 Replies Latest reply on Apr 4, 2017 6:59 AM by Anthony Claypool

    Using Calculations to match records to track change over time

    Anthony Claypool


      I'm a data viz rookie, so it's entirely possible that my assumption that the solution is a calculation is incorrect.  I would attach a workbook, but...FERPA.  I think I'm describing it ok down below, so maybe some amazing person will be goodly enough to give me their thoughts. 


      I'm trying to explore, how to use Tableau to support longitudinal study of student achievement with a really tall data set.  If I made the dataset wide it'd be an obvious solution, but that's a decent amount of curation I would rather avoid if I can.  So for example, we have tens of thousands of student data records over 5 years (three terms per year) of student assessment data.  I'd like to see how their percentile has changed over time.  The aggregated view (AVG) is easy, but trying to use calculated fields to track and individual student's work over time (with thousands of different students) has been a challenge.  My impulse is to write a calculation that filters by [Term Name] then subtract [Percentile] or [Test Score] from a value from a different filtered [Term Name]. 


      I tried tinkering with the Case function, but it was null-city. 


      I'm thinking there needs to be some sort of match function in the calculation.  General thinking Filter->Match->Compute. 


      Any advice out there?