2 Replies Latest reply on Feb 2, 2011 8:42 AM by philmodjunk

    Accounting for human error



      Accounting for human error


      I'm looking to write a script that would look for mismatched information.  I'd like it to perform a find and show only the records where the data in a given field is the same (i.e. all the names that are in the database more than once).  The information in other fields won't be the same (John Smith is there on 1/20 and 2/1, I want to check that his phone number is correct for both entries.)  How can I do this?  Thanks!

        • 1. Re: Accounting for human error

          Can I side-step the question for a bit, and ask what creates the possibility in the first place?  is the reference to 1/20 and 2/1 two dates, maybe that John Smith was 'there'?  Why would you ask someone to create two valid entries (I assume) for John Smith, and expect them to enter his phone number (or zip code, or employer, or whatever) each time?  If you create a table with all of John Smith's details in it once, and relate this table to that one, then they don't have to enter the details twice, and they can't get it wrong.

          • 2. Re: Accounting for human error

            To fix the current problem, you need a single field that contains all the data that might identify two records as duplicates of one another.

            If both first and last names are entered into the same field, you have the field you need. If not, define a calculation field with a calculation such as: FirstNameField & " " & LastNameField.

            Then enter find mode and put an ! in that field. This is the duplicates operator.

            After you perform the find, sort your records by this field to group the records by each duplicate value so that you can examine and compare them.