4 Replies Latest reply: Mar 10, 2016 7:25 PM by Fara Naz RSS

    Set Class Size

    Fara Naz

      How can we use datawatch to define Class Size based on data? In example I have set of data that store number of expenses per peron on daily basis, so I would like to generate a table that group the expenses amount according to ranges (range example: 1000-2000 dollars, 2001-3000 dollars) but the ranges will be determine by the min and maximum value of the data and it should have 10 group of ranges only.. can any body help me

        • Re: Set Class Size
          Theo Klemming

          Hi

           

          It sounds like this is a perfect case for BUCKETING; On your expenses column, right-click and say New Bucketing Column. Then select 10 equal distance or equal density buckets. Which one you use depends on your preference: Do you want 10 groups where the limits have the same distance, or do you want 10 groups that have the same number of persons? Equal density is what gets you the same number of persons/members in each group.

           

          Also, take a look at this video:

          https://www.youtube.com/watch?v=GCSm_WwuPhw

           

          best regards, Theo Klemming

            • Re: Set Class Size
              Fara Naz

              thanks theo for the response, i would like to have 10 groups with same distance, i did use the equal distance type with no of bucket : 10, but it returns only 5 buckets.. what else that I missed? please advice

                • Re: Set Class Size
                  Theo Klemming

                  Hi Fara

                   

                  Have you included the new Bucketing column in your Breakdown on the visualization? You have to in order to see your data organized by each bucket. Also, in case you are using a numeric column on Size or Height, which for some buckets amounts to zero or something very small, you may not see those buckets (groups). Lastly, if your data has only 5 data rows, you will not be able to get more than 5 buckets. If you can share your data set with me, I'd be able to provide additional advice. My email address is theodor <underscore> klemming <at> datawatch dot com.