Direct Extraction Vs Field Manipulation

1 post / 0 new
Brian Element's picture
Brian Element
Joined: 07/11/2012 - 19:57
Direct Extraction Vs Field Manipulation

Hello Group Members,

Direct Extraction is applied in IDEA to extract specific records from a parent database into a child database which meets a specific condition or conditions.
As an example Direct Extraction can be applied on a Payroll file to capture (extract) all Payroll payments where 'Overtime' has been paid. Upon applying the Direct Extraction a separate child file gets created in the IDEA File Explorer below the Payroll file with all 'Overtime' payments. Now think of some other indicative rules like 'Salary Arrears', 'Nil Withholding Tax' and more. Each of these rules create separate child files in IDEA. At the end of the data analytic assignment you end up having as many child files in IDEA as the rules you tested for. This could lead to a lot of collation time to get a uniform picture.
To save on this time you can try Field Manipulation. Field Manipulation allows you to append newly computed fields right into the parent file without having to create multiple child files.
So taking the Payroll example cited above you can append a virtual character field titled 'Overtime Remark' with a criteria @if(Overtime=0, "No Overtime Paid", "Overtime Paid")
For 'Nil Withholding Tax' you can append a virtual character field titled 'Withholding Tax Remark' with a criteria @if(Withholding Tax = 0, "No W Tax Deducted", "W Tax Deducted)
Appending fields through Field Manipulation lets you have all the rule based testing conditions in the form of Comments in the parent file itself. This makes review and collation easy since its all in one file.
Further if you would like to score your rule testing with 1 for Non Compliance / Exception and 0 for Compliance / No Exception you can append a virtual numeric field titled 'Nil Withholding Tax' with the criteria @if(Withholding Tax = 0, 1, 0). This will append a 1 in the payroll line items where W Tax has not been deducted and 0 where it has been deducted.
Now imagine you can assign such score for multiple rules in separate newly appended fields in the parent file and then just add up the 1's to get a Non-Compliance / Exception score.
Best Regards
Group Admin