Ron, Thanks for your detailed response. (I have copied, below, the text from the archive, since I did not receive the e-mail.)
I should have double checked the options to notice that surnames must match all characters. (I have never tried the Sounds Like option, due to negative experiences with that in other search engines. Too many hits.) I now see that my two Isaac Walkers would fail the match because their birth dates are more than 180 days apart. I still do not see why my two Hannah Walkers would fail to match. The spouse surnames do not match, but it is still my contention that that should not be part of the criteria. The birth dates were 1883 and 26 Feb 1883, which should match, I think. I wonder if '1883' resolves to a specific date, like 1 Jan 1883 or 31 Dec 1883, for purposes of the matching? I'll have to do some testing, but I think I will change the birth date match from 180 days to 366 days. Regarding your suggestion to find duplicates, then tweak the options and check again: this is not an available technique during the drag & drop merge process, where the duplicate search happens only once, and only on the individuals being merged. Certainly I could follow up with a file-wide duplicate search. I had been avoiding this, since I have 122 potential duplicates to wade through and deal with. Most can immediately be marked as not duplicates, but a few are ambiguous and require a bit of research. I guess I had better do it. Ward ----------------------------------------------- Ward, Note that in "Normal Duplicate Search" that the Surnames are compared as either "Exact Spelling" or "Sounds Like". It is the Given Names that are compared for the number of beginning characters and those have to match exactly...there is no "Sounds Like" for the given names match. Set the "Check birth dates, Day range" to 0 for an exact match or some other number of days that might find differences. The more criteria you specify, the fewer results will be found. Un-check "blank birth dates", "birth locations", and all the other date fields so that only birth is being considered. Take into account "Compatible Gender", "Ancestral File Number", "FamilySearch ID" if you wish but almost always un-check "User ID". Set the "Compatible Parents" and the same parameters you used for the individual. Click "Continue" Resolve any possible duplicates that are found. Adjust the parameters to use the death date and blank the birth date. Resolve any more possible duplicates. Adjust the parameters to use the burial date and blank the death date. Resolve any more possible duplicates. You should have a fairly clean file now. Additionally you may want to look at the Name List sorted by Surname and resolve any possible duplicates you see there. Lastly, look at the Marriage List sorted by Husband, resolve duplicates and then sort by Wife and resolve duplicates. Now only those really hard ones will still remain. Others may have some tricks they've developed for finding additional duplicates. Then you might want to tackle duplicate marriages, duplicate events, duplicate sources, duplicate pictures, etc. It's all fun. Ron Taylor -----Original Message----- From: Ward Walker Sent: Saturday, June 14, 2014 1:30 PM To: [email protected] Subject: Re: [LegacyUG] Find Duplicates does not handle middle initials Well, with some digging, it turns out that my example was not valid, but I think I have some other examples that are. In my example, the actual missing match was with Ruth Emma Cook and Ruth E. Cooke. So the surname did not match within 5 characters. I now spot the following duplicates in my file, which were brought in via the drag & drop merge of the mothers line: - Isaac J. Walker and spouse Angelina Carmon; and Isaac John Walker and spouse Angeline Carmon. - Hannah Ann Walker and spouse Joseph Farquhar; and Hannah Anna H. Walker and spouse Joseph Farwuhar. The two Isaacs have births one year apart. The two Hannahs have the same birth year. They all have the same parents.` I see no reason why the Isaac/Angelin* couples were not presented as potential duplicates. For the Hannah/Joseph couples, the spouse's surname does not match within 5 characters. Certainly the parents could have had two different Hannahs, although it would be unlikely that both lived to marriage in that case, and they do have the same birth year. (One has a month and day as well.) Is it a requirement that the spouse must also be a potential match, for this to be presented as a potential duplicate? I've had way more far-fetched potential duplicates presented to me, including twins with different given names. Ward -----Original Message----- From: Brian/Support Sent: Friday, June 13, 2014 1:41 PM To: [email protected] Subject: Re: [LegacyUG] Find Duplicates does not handle middle initials If that is the case do your Ruth E. and Ruth Emma have the same husband in both files so they can be considered due to family relationships? Brian Customer Support Millennia Corporation [email protected] http://www.LegacyFamilyTree.com Legacy User Group guidelines: http://www.LegacyFamilyTree.com/Etiquette.asp Archived messages after Nov. 21 2009: http://www.mail-archive.com/[email protected]/ Archived messages from old mail server - before Nov. 21 2009: http://www.mail-archive.com/[email protected]/ Online technical support: http://www.LegacyFamilyTree.com/Help.asp Follow Legacy on Facebook (http://www.facebook.com/LegacyFamilyTree) and on our blog (http://news.LegacyFamilyTree.com). To unsubscribe: http://www.LegacyFamilyTree.com/LegacyLists.asp
