Earn a 50% discount on the DP-600 certification exam by completing the Fabric 30 Days to Learn It challenge.
I am doing analysis of duplicate entries for a survey. I want to analyse the duplicates to determine which one to keep before deleting the rest of the entries.
I was able to learn how to group the entries by the unique staff ID, and using the count row function i got this screen below:
My next step was to click the individual entries (example: A0612) to see all the duplicate entries. In the example below, I have one unique staff ID with 8 entries. This could have been because the staff submitted multiple survey responses.
Question:
1. I have over 30 columns. How do i quickly analyse the similarities of row entries? My initial thought was to unpivot the other columns, and compare the values for each attribute column headers. But is there an easier method than this?
2. After I have decided which duplicates to delete, how do i ensure these changes get reflected in the master sheet? The long method would be for me to manually delete each row one by one. Is there a faster way around this?
You can use Table.FuzzyGroup to accomplish this. This function will group similar column values together.
Start by unpivoting the data into attribute/value pairs (Question/value)
Then do a group on all the columns with a count rows
Finally, replace the Table.Group function with Table.FuzzyGroup function.