* Quality of data

[Gowermick]

One can easily allocate a quality rating to any fact!

How do others rate the quality of esoteric things such as parentage?

When one finds a 'potential' christening of an individual, by extension, the names of their parents are also 'potential'. How do others mark this relationship as potential, and worthy of further investigation?

The only way I can think of achieving this is to create a flag. Has anyone got any other suggestions?

[tatewise]

To be specific the quality rating is allocated to a Citation in its Assessment field.

So you can set that rating for any item that can have a Source Citation.
Such items certainly include Facts (Events & Attributes) but also Names and Places and Witnesses.

Therefore, as well as citing the Christening Source from the Christening Event, also cite it from the parent Names, and set the rating in each Citation Assessment.

It is more difficult to set a rating on family relationships such as Parent-Child and Spouse, but here is a method.
Each relationship has a Parents family link and a Spouse family link that can be seen on the All tab.
Those links can have a local Note which could specify the rating.
(Right-click the Parents family or Spouse family link and choose Add Note > Add Note to this Record.)
The Note is normally only visible on the All tab but can be customised into the Main tab.
Use the Menu > Customize Data Entry and add items with Label and Data Ref such as:
Parents 1 = %INDI.FAMC[1].NOTE2%
Parents 2 = %INDI.FAMC[2].NOTE2%
Spouse 1 = %INDI.FAMS[1].NOTE2%
Spouse 2 = %INDI.FAMS[2].NOTE2%
Then you can set the rating directly on the Main tab.

[Gowermick]

Mike,
Your suggestion to qualify the relationship by adding a note and qualifying that, is certainly one way to go, especially if one could then use this in a filter for a query.
Thanks

[mjashby]

Mike,

If the initial evidence is tentative, I tend to go for creating an entirely separate family group for further investigation and if satisfied, at some later date, that the two individuals are, in fact, the same person it is relatively easy to merge those two records and thus draw in all of the extra connected relationships. Equally, should the two individuals prove to be different, it is also relatively easy to delete/sideline any rogue family grouping together with the documented Source Records and linked Media.

There is a specific GEDCOM Tag - ALIA (Alias) which you can use, should you want to create 'formal' links in the GEDCOM File between potential matches. This Tag was/is specifically intended for linking two or more separate Individual Records in a GEDCOM File which may prove to relate to the same person. See - http://wiki-en.genealogy.net/GEDCOM/ALIA-Tag for a further description/explanation of syntax and usage.

The potential problem I find with 'hard-linking' possible/probable matches from the start is that, if the evidence for this one relationship starts with a best possible or most likely match, but one which still requires further corroborative evidence, then it is almost inevitable that you will have to build a wider range of family relationships (more people) who may or may not eventually prove to be part of the primary family group. Once those relationships have been asserted formally it can become problematic to disentangle them, or even to convince yourself and/or others that the original assumption was probably incorrect. - People perpetuating incorrect assumptions by auto-merging with Public Tree information on Ancestry and other sharing sites spring to mind as one extreme example of convenient matches become accepted as reality, even if/when the originator of the original assumption(s) has later identified their original error(s) and corrected their own research information.

Mervyn

[LornaCraig]

Tatewise wrote

It is more difficult to set a rating on family relationships such as Parent-Child and Spouse, but here is a method.

As an alternative to the method Mike suggests, you could use the method suggested in the notes on "What's New in Version 6.1", below. Then set an assessment level on the source citation for the witness role. But whichever way you do it you still need some way of seeing "at a glance" that the relationship is uncertain. If I am not attaching any other relatives to the probable parents then I find the simplest way of indicating the uncertainty is to add (?) after their names, as in John(?) Smith. However if you have other relatives linked to the parents then Mervyn's approach of using a separate family is safest.

[AdrianBruce]

One caveat I always try to make about potential parents is that you need to distinguish between these two cases:

1. I think that the father of John Smith is William Smith (1792-1852)
and...
2. I know that the father of John Smith is William Smith (e.g. from John's marriage certificate) and I think that that William Smith is William Smith (1792-1852)

The problem is that there's an awful temptation to summarise case (2) as case (1) - I've done it myself - but to do so loses information. Not so much a problem if you're writing out stuff in notes, but the more you formalise it, the easier it is to lose information if you try to make things simple.

[Gowermick]

Thanks all for your suggestions. Apart from Mike's note and my Flag, the other suggestions seem a bit too heavy for what I want.

Mervyn, I take your point about adding potential relatives, who than have to be deleted, but by unlinking the child, all associated relatives then become unrelated to the root, so can easily be filtered and removed, especially in FH6 (sadly I've experienced this situation before ). Every once in a while, I do a spring clean and check and confirm all relationships. One would be surpised how much incorrect data still manages to slip in! Painting Forth road bridge springs to mind.

As for trees on websites, my on-line tree carries a warning : data may be inaccurate, you are free to copy it, but you do so at your own risk!
As for copying other people's trees, I always treat them as hints, just the same as FH6 internet 'hints', and don't use/copy the information until I'm happy with it.