Genealogy from the perspective of a member of The Church of Jesus Christ of Latter-day Saints (Mormon, LDS)

Tuesday, April 7, 2015

What is the difference between a combined record and a merged record?

The question in this blog title was sent to me by email. I realize that the difference between the two types of actions can be confusing.

The idea of "combining" rather than merging records started with the introduction of the program (NFS). I should note that sometimes the program's title is written "" or (nFS), but either way, both titles refer to the same program. NFS has been taken off line and I can no longer show screenshots from that program. The idea behind combining the records was to preserve every single variation of the copies of the record for view. The duplicate copies came from the same individual being submitted multiple times in the programs that were used to seed NFS.

There were originally five large databases that were combined to form NFS: the Ancestral File (AF), the International Genealogical Index (IGI), the Pedigree Resource File (PRF), the Church membership records and the Temple records. Although it was possible that there were some unique individual entries, putting all five of these huge databases together resulted in some individuals having hundreds (over a thousand?) copies. When the NFS program was introduced, the obvious duplicates were combined into one record. There were, however, many less than obvious duplicates that were not combined. The program had a link to view the variations in the list of combined records. Unfortunately, since not all of the possible copies of an individual were initially combined and after the program was introduced, the users were allowed to find and add further duplicates to the combined records. Inevitably, there were errors made both by the program and by the users in combining records that were not duplicates. There was also a procedure for uncombining records.

Eventually, as NFS developed, there was a problem with the total number of combined records. In many cases, the number of duplicates exceeded the ability of the program to adequately process the information. An arbitrary cap was placed on the number of combined records allowed. This number was, in some instances, far less than the number of actual duplicates in the database. As a result, this left a huge number of uncombined duplicates of some individuals. These individuals came to be called "Individuals of Unusual Size" or IOUSs. These individuals were primarily those who had joined The Church Of Jesus Christ Of Latter-day Saints before 1900.

It became apparent that the NFS program was unworkable. The mechanism for "combining" records created a great deal of confusion and the fact that a cap had been placed on combining the existing duplicates created a potential for even more duplicates in the program and also created duplicate Temple ordinances being done for duplicate individuals that could not be combined. Combining the records preserved all of the duplicates in the program.

As a result, FamilySearch began the process of transitioning to a new program called "Family Tree." This program was intended to rectify the deficiencies of Without going into a detailed analysis of the differences between the two programs, it is sufficient for the explanation concerning the difference between combining records and merging records to explain that Family Tree abandoned the concept of keeping all of the combined records and instituted a process of "merging" records. The idea of merging records is that only one record survives the merge process. Information from the merged record is preserved in the final version. In actuality, none of the information is lost because there is a process of unmerging the records.  However, once the duplicates are merged only the surviving record is visible in the program. Presently, the limitation on the number of combined records, a holdover from, limits the number of merged records. The reason for this limitation is that information is still being transferred from to Family Tree.

The final results of the introduction of a merging process is that "combining" is no longer available. In Family Tree records are merged not combined. When the process of transferring all of the data from to, Tree is complete the limitations on merging records will be eliminated and the rest of the orphan duplicates can be found in merged. The ultimate goal is to have only one individual in the program for each individual ancestor. The mechanism for doing this is present and will be fully available when the transfer of information is complete.

I realize that this is a somewhat complicated explanation but the idea is that we only have to be concerned about merging records presently. Combining records is no longer available.


  1. This is how I have been teaching others the difference between combining and merging.

    Combining was like adding a link to a chain. Each time the record was combined with another record, it became a longer chain of information. Imagine a heavy long metal chain.

    Merging - takes two records or chain links and merges them into one link. Every time you merge, you still have only one chain link.

  2. I like that Amy. A side issue here is how Family Tree lets you find duplicate records that need to be merged. The Possible Duplicates tool is the weakest and only finds close matches. That is what is being used on the impedance for reserving names for the temple now. The Find feature lets you see more possible dups, And then Search Records shows more than Pos Dups tool to. Also, if an unmerge is not done within a few months, I understand that the record left on the right is tombstoned, and eventually not recoverable. Always remember that temple work is kept on entirely different data bases associated with each temple. So that info is not lost in a merge.

    1. Thanks so much for the additional info.