5 Replies Latest reply on Feb 27, 2012 4:23 AM by Sprucegum

    Entity Relationships - Confused!

    datamarine

      Title

      Entity Relationships - Confused!

      Post

       I've just finished reading 'The Missing Manual' for Filemaker Pro. I know a great deal more than I ever thought I would, but I'm still completely confused about the relationship diagram that I need - of course no example can ever mimic exactly what is required, but because all the examples I've found seem to relate to businesses with clients, invoices or stock, I remain completely lost and unable to translate what I've learnt into anything that remotely resembles what I need. Mine is a research database - let's say I have a person whose address changes frequently, who marries and has children (and exactly when and where those things occur is vitally important). I am befuddled about how on earth I can translate the examples for clients, invoices and stock into anything that comes anywhere remotely approaching my particular problem. I'd be very grateful to hear from anyone who has used Filemaker Pro (v10) to write a relational database. Or anyone who has any suggestions that might help, of course.

      Best wishes, and thank you in advance.

       

        • 1. Re: Entity Relationships - Confused!
          Sprucegum

          Hi Datamarine,

          Below is an example of 3 related tables matching your description (if I can figure out how to post an image).

           

           

           

          • 2. Re: Entity Relationships - Confused!
            datamarine

             Hi Sprucegum

            Thank you so much for your kind help. I am much clearer in my mind about how to set about it after looking at this so it was extremely helpful. I hope I am not taking too great an advantage of your generosity by asking another question! For my purposes, I am only following the males, many of whom will have married more than once and ended up with a large family. Because of my research I am not interested in couples per se - the females only matter because they are the mothers of children (and the place of marriage is important).  Would you recommend in that case, having the ID just for the male, (not the couple), adding an extra table to show spouses, but adding children to the main children's table irrespective of who the mother is? Or would I simply replace the couple table in your example, with a male ID table (which could include where and when they married). Sorry if this is a no-brain question!

             

            • 3. Re: Entity Relationships - Confused!
              Sprucegum

              Hi Datamarine,

              I think you're probably right. Given your description, it sounds like you need the following tables; Fathers (including only your subject population of adult men), Addresses, Marriages & Children.  Note the Marriages table stores both Marriage & Spouse info.  Also, (anticipating your need to run statistical analysis on your subject population of Fathers) it might make more sense to create a separate table for Children. Below is a revised diagram.

              • 4. Re: Entity Relationships - Confused!
                datamarine

                 Hi Sprucegum

                What can I say? Thank you! This is terrific - thank you for confirming that I was at least on the right track, though I'd never have been able to sort this out in a million years.  I'm glad of your suggestion for the separate table for the children. I can't thank you enough for your generous help, and your patience explaining that to me. I'm really very grateful!

                • 5. Re: Entity Relationships - Confused!
                  Sprucegum

                  Hi Datamarine,

                  Looking at this after my AM coffee, it occurs to me (if the play-writes are to be believed) you may need to allow for the probability that some children are born to women the father is not married to and may never live with;)

                  From a genetic stand point, a fathers' residence or marriages are probably not as relevant as birth parent identities or where children are reared.  From a social and theological stand point it may be interesting to study the relationship between marriage, promiscuity and human migration.  And, of course, children may be adopted or given to the legal custody of non-birth parents.

                  I've provided some examples, but you'll have to consider carefully the nature of your research and how to structure data tables to best suit your needs.

                  Best of luck