Follow
Tree Repairing
- how to correct basic errors on the tree.
Because this is a tutorial project please do not link profiles to it. Translations are welcome.
If you have any specific questions, comments or suggestions about the guidelines below, please contact June Barnes or Erica Howton who are Volunteer Curators. Do not hesitate to ask questions about anything that is not clear, or draw attention to things that I may have been missed!
Alternatively raise a discussion to embrace wider participation and input.
See the following user tutorials:
- Merging - Step by Step guidelines to merging duplicate profiles.
- Merging - Case Studies and Examples
- Searching
- Matches
- Data Conflicts
- Tree Conflicts
- Basic Techniques 1. Viewing Preferences
- User Tutorial - Basic Techniques 2. Actions Menu
- Basic Techniques 3. Profile Tabs
See also
Please note that profiles used in the screen-shot illustrations have no significance and are merely used to illustrate scenarios - they have no relevance to the tutorials. The illustrations will not always match your own - this depends on how you have set your Name Preferences, and what your choices are when viewing the tree.
Occasionally I use a "vertical tree layout" in order to cover more ground.
How much of what is explained you can carry out depends on whether you have a Geni Basic or Pro subscription.
There are often situations in the tree where there is a very basic error which needs repairing. These are instances where you are unable to merge duplicates from a stacking tray and get a message such as Cannot merge this profile with his father, Cannot set parents, or a reference to Splitting the tree.
The circumstances listed below suggest that something is wrong. It is usually best to cancel the stacking tray and investigate the tree first before selecting one set of parents for instance over another. It could be resolved by simply changing a relationship, (See Tree Conflicts, but often it requires a merge to be undone. Check the revisions tab to see if there is anything that looks like a bad merge.
Geni Tips offers solutions to some of the simpler scenarios that you may encounter as you build your tree, e.g. adding a spouse as a sibling, adding a sibling as your parent etc.
The following circumstances may indicate that the error is perhaps more complicated and suggests a deeper, more far reaching problem. Most are solved by moving or changing relationships, some are a little trickier. In time we hope to bring some examples to the table.
You may notice that ...
1. There are half siblings or step children in the family that should not be there (See C2 Oosthuizen below).
2. Two sets of parents
If you have a stacking tray showing two pairs of parents, selecting the "correct" pair may just cover up a problem. There is usually a reason for that showing up - the most likely being that there has been a bad merge. If the "correct" parents are chosen then the "abandoned" couple end up minus a child of that name - their child effectively being merged into the other. Just changing the parents either in the stacking tray or through the relationship edit page effectively loses a child in one family. The best course of action in such instances is to look at the trees of both profiles to see firstly if there are two distinctly different families where that profile is not a duplicate and should not be merged - therefore get a curator to undo it, or secondly whether they are indeed the same family/person and one tree has incorrect parents in place. This is touched on in the 4th case at Merging Case Studies - bottom TC4
3. Parents are incorrect or missing (known to be there before).
4. Missing Spouse which you know was there before
5. There is a wrong or extra spouse
- See C3. Prentice below.
6. An extra generation has been entered (e.g. a profile has been added as a child of themselves).
7. Profiles are placed in the wrong generation - up or down the tree
8. Two fathers (perhaps adopted).
Theunis Botha is a good example of adopted parents. See How can I add adoptive/foster parents to the tree?
9. Children who were born after the death of their mother.
10. A mother born after the birth of their child.
Worth mentioning here to show how - A person has two relationships within a family - see - Multiple ways within the same blood tree. If you have two people who are related in multiple ways within the same blood tree (i.e. first cousins who are also married), we call these types of relationships "Cycles" on Geni.
Often the easiest solution you might consider is to delete the incorrect profile and re-add it, but this is only to be done in situations where you are the only manager involved.
If other people also manage the profiles involved then it is NOT acceptable to delete the profile.
You need to contact the managers to explain the problem, inviting their co-operation in resolving the issues. There is often an easy solution - changing a relationship, merging or moving - if there isn't then you can ask for help. Curators have the expertise and tools to help in difficult situations.
More examples will be added as they arise. We learn from such cases that careless merging causes repercussions which are often hard to undo. Curators have only recently had the ability to undo merges, so before August 2014 it is not an option, and it can be almost impossible to find the root of the problem. Sometimes the only answer is to re-build.
Note that often the information to help make a correct merge is already available in the profile. People add documents which explain or source information - check under Media and Sources. People often make notes in the "About Section" - read them if they are there. If a curator has added a note to the profile page - read it - it may have vital information!!! It doesn't take that long to look a little further before accepting a merge which seems to be "not quite right" and saves a lot of time in the long run.
Case 1. Maria Rousseau
The profile for my 8x great mother Maria Rousseau (Figure C1:1) raised alarm bells because the relationship path was very much out of kilter. Being familiar with this line of my ancestry I knew that the name of Botha was out of line. In addition her husband, Jan Janse van Eeden and all her children were missing.
Figure C1:1
In order to correct the tree I needed to establish firstly what the correct path should be, and secondly how things were made to be incorrect. This usually involves bad merges which need to be undone if possible.
I used my personal records, backed up by the details at First Fifty Years Project (Figure C1:2). to establish what I was working towards.
Figure C1:2
My first step was to look at the revisions on her profile. I took screen shots of the lists for reference and noticed that there had been a large number of merges carried out over the past year. I systematically undid them working backwards from the present time, keeping a log of what I did. I kept a record of the profile merges I undid with links and screen shots of the duplicate profile pages. I made notes as to their accuracy, noting if they were good matches so that after I had gone through potential problem merges I could return and remerge them.
The fourth one I undid seemed to be the culprit. It was merged on 11th September 2015. Figure C1:3. The relationship showed a link to "Botha" - the same relationship which was noted as incorrect in the first place.
Figure C1:3
Having done the 4 un-merges the tree for Maria still didn't look right. (Figure C1:4). However one of her children, Frederick, was now in the picture, but his father was not named. There was a tree conflict on his profile which I explored further.
figure C1:4
In Frederick's stacking tray, (Figure C1:5), it looked as if the two fathers were a match. Before merging them I went to his profile page (Figure C1:6) to make sure that there weren't any problems to fix before merging.
Figure C1:5
Figure C1:6
Figure C1:7 His profile page (C:1:6) reflected 3 mothers and 2 fathers. I explored this tree further and found that one of the duplicates for his mother was a match and processed that merge. The other Maria Rousseau was the one in C1:3 above. Looking at her tree (Figure C1:7), revealed that she was connected to Theunis Botha as his wife. Research established that she was in fact his mother-in-law, and I removed that relationship and merged her with the main Maria profile in order not to loose the manager of that tree from the profile. Not doing so would leave her as a "floater" on Geni, unlinked to anyone. I processed the obvious merges resulting in Frederick's tree as in Figure C1:8.
Figure C1:8
I explored the conflicts further, merging as necessary. Frederick's tree now looked healthier, (Figure C1:9), but still reflected some problems and conflicts. My own research had not establish the names of any parents, something which needed to be substantiated, and there was still a tree conflict for Maria.
Figure C1:9
The stacking tray, (Figure C1:10), gave rise to further research into the two mother's profiles and the existence of a sister which she was not known to have.
Figure C1:10
It seemed that the two mothers were the same person in spite of the minor differences, but the names of her parents were different. Because it is understood that Pierre and Maria were siblings I arranged them so that they both had the same mother and raised a query regarding the authenticity of them as the parents of Pierre and Maria, and also querying the names of Maria Retif's parents. Having made all the changes catalogued above Maria's profile and tree appear as below - C1: 11 and 12.
Figure C1:11
Figure C1:12
Case 2. Oosthuizen
My attention was drawn to the following profile for Elsje Oosthuizen which reflects that she had a half brother. Figure C2:1.
Figure C2:1
When I viewed the tree, (Figure C2:2), I saw that his mother had an un-named spouse with a child Gerhardus.
(Incidentally the yellow borders to the profiles is the result of pressing the CMD button whilst taking a screen shot).
Figure C2:2
The simplest solution would be to move him into the family, allocating Johannes as his father. This I know to be the case as I am familiar with the tree, and which I also confirmed from my personal records and other reliable sources. However, he has a wife "Anna Botha", which my sources and the rest of the tree shows is not correct. The duplicate for Gerhardus in the main body of the family shows his correct two wives.
On the tree we can see that there is a green orb with x2 in it, (Figure C2:3), indicating that Anna Botha appears twice on the tree. I followed the green link and found her as the wife of another son in the family. (Figure C2:4). This according to the records is correct.
Figure C2:3
Figure C2:4
Investigating further on his profile page I saw that a profile for Gerhardus had been merged with Jan Hendrik Oosthuizen. (Figure C2:5). This merge had been undone but I had not followed through (tut tut!), - easily overlooked but no excuse! I removed the relationship between Gerhards and his sister-in-law.
Figure C2:5
There were a number of ways to place Gerhardus in his rightful place in the family. For the sake of this illustration I am working through all possible avenues. Figure C2:6 shows the tree after the removal of the incorrect wife.
First method - if the blue orb was for a Smart Match the merge could be completed that way.
Figure C2:6
Figure C2:6
A Second possibility, once the incorrect wife was removed, would be to "drag and drop" the father onto the vacant "add this person" node using the CMD button. Figure C2:7. See Merge Guide for an explanation on how to do that.
The third possibility - use "drag and drop" to move the profile to the correct position - Figure C2:8.
You could also use the "move" option from the actions available on the node - explained in the Merge Guide project
Figure C2:8
... which takes you to the comparison which you could accept and merge. Figure C2:9.
Figure C2:9
Accepting that merge left the tree as in Figure C2:10
Figure C2:10
The stacking tray, (Figure C2:11), shows there to be no other duplicates to merge. Clicking "assign parents" takes you to the options available. (Figure C2:11). You could select the correct parents, (the first on the list), and complete the correction.
Figure C2:11
Figure C2:12
The fourth method would be to change the parents in the relationships edit window. Figures C2:13, (upper section of the window), and C2:13a (lower section of the window), show the relationships window for Gerhardus.
Note this is not available to Basic users.
Figure C2:13
Figure C2:13a
Ticking the box alongside the first relationship listed in Figure C2:13a, followed by clicking "Remove" and then "Save" will remove that relationship. (Figure C2:13b)
Figure C2:13b
By which ever method employed the conclusion to the exercise is as in Figure C2:14.
Figure C2:1
Case 3. Prentice
The following showed up in my Tree Conflicts.
Figure C3:1
Robert Prentice was linked to a 2nd wife and children and there was also a Tree conflict on his son Thomas. This situation was not correct, confirmed from a little research and background reading. I established that he was not married to Elizabeth Stuart, and so although it would be easy enough to disconnect him from her it was clear that an error had been made linked to Thomas.
The stacking tray for Robert is shown in Figure C3:2....
Figure C3:2
... and for Thomas in Figure C3:3.
Figure C3:3
Thomas's profile page, (Figure C3:4), reflected the multiple parents.
Figure C3:4
Checking the revision history showed that there had been 3 merges which could have caused the problem, and which could be worked through to establish if any were mistakes. Figure C3:5.
Figure C3:5
I next looked at the profile and revision history of his father Robert, which showed the multiple parents and the two wives. Figures C3:6 and C3:7. There was only one merge that could be the culprit.
Figure C3:6
Figure C3:7
I undid the merge with Robert Prentice done on the 5th July. Looking at the profile that had been merged in for Robert Prentice (Figure C3:8) I saw that he was married to only Elizabeth Stuart.
Figure C3:8
There were two matches (blue box) which I explored whilst I was at the profile - Figure c3:8.
Figure C3:8
Figure C3:10
Both were not matches (one listed was the one which I had undone), and so I removed them.
Having undone this merge I returned to the tree view to see that the family was now restored to how it should be, (Figure C3:11 and C3:12), and that there was no more work to be done.
Figure C3:11
Getting Involved
Feel free to follow, request to collaborate
Further Reading and related projects
Useful Links
Geni Start here Basics
User Guide for Merging Profiles by Eugene Thomas
How to merge profiles on Geni.com Video by Randol Schoenberg
Working sensibly - The ABC of merges, merging and linking profiles is a project which explains how to check for profiles before tree building by searching for them first.
International Help Portal
Working with the Big Tree
Connecting to the Big Tree