Source Consistency Issues
I encounter a considerable amount of source consistency issues, especially with census data in England. Many of my ancestors were farmers were tied to the land and did not move. When completing the census sheets the taker enters the civil or ecclesiastical parish or other administrative district. When contributors copy that information as a residence, the algorithm catches it and notes it does not agree with the abode. A prompt box might be necessary for contributors fully open the source and view the image, if available.
Commenti
-
What I see as a problem is that very few people rename source titles from the generic name given by familysearch.
1 -
I think that the main problem is, that the "Quality Soring" is being done by Americans, who have little or no knowledge of British Genealogy.
0 -
Thank you all for your input. This is exactly why we need your help :)
0 -
"No History, "West Virginia, Deaths, 1804-1999" has a last name of History, which is different from Love."
here is one of the ridiculous things in Quality
1 -
@Mormor192 can you provide a PID for the profile where you are seeing this issue?
0 -
Here's one example of this: https://www.familysearch.org/tree/person/sources/K2QM-W23
This is semi-hillarious. Looks like it was the practice for death certificates in West Virginia at the time that if a parent's name was not known, then that unknown parent was listed as "No History." And when the death certificates were indexed, they strictly followed the practice of indexing exactly what was on the record.
As I've mentioned elsewhere, it seems the Data Quality Score is going to have the unintended consequence of pointing out far more errors in the indexes than it ever will on profiles.
You can see in this search: https://www.familysearch.org/search/record/results?count=100&offset=200&q.givenName=no%20history&q.givenName.exact=on&q.surname=%2A&f.collectionId=1417434 that there are about 660 records with one or both parents listed with first name of No History in this collection. Searching for last name of No History: https://www.familysearch.org/search/record/results?count=100&q.givenName=%2A&q.surname=no%20history&q.surname.exact=on&f.collectionId=1417434 gives another 235. Removing the constraint of being in the single collection for West Virginia gives a total of 1685.
It's starting to look like there needs to be a very efficient way to dismiss conflicts that are due to problems with an index.
1 -
@Gordon Collett Thank you!!
0 -
Oof, "semi-hilarious" is right. (You gotta laugh, because else you'll cry.)
But those indexers weren't following instructions: all of the projects I've ever worked on have included something along the lines of "if the name was entered as some variation on 'unknown', mark the field blank." "No History" is clearly a variation on "unknown", so all of those mother's names should have been marked blank.
2 -
@Gordon Collett your search query is very helpful.
When we can identify a record collection with the same problem repeated, it is more likely to get fixed.
I have created a ticket Hundreds of relatives named "No History" in [West Virginia, Deaths, 1804-1999] (1417434)
2 -
Not certain what your PID means. And no, I can't now, as I've looked at too many in the past few days. I lucked out in being able to post a screen shot yesterday or Thursday, but no luck in trying to do that today.
There is one where the husband or person writing down or transcribing the information gave what was probably the length of illness instead of the date of birth , thus making a married woman two+ years old instead of her actual age.
0 -
Here is the Ma
son County is 98 miles from Mason county absurdity:
0 -
Wow, that's a fun one, @Mormor192.
There are two different counties in the Places database that are labeled as "Mason, Virginia, United States" for one of their time periods:
today's Mason county, Kentucky, for the period 1789-1792 (Place ID 394366), and
today's Mason county, West Virginia, for the period 1804-1863 (Place ID 393605).What that message is telling you is that at least two conclusions on that profile have had their locations standardized as one of these counties, and at least one conclusion has had its location standardized as the other one.
2 -
The person in this example had no contact with the Kentucky Mason. If someone with no connection to the family put them in the wrong Mason and it isn't obvious from the data, I really can't do anything about it or worry about it.
0 -
Fortunately, we now have the data quality routine that does tell us when there is an inconsistency in the data and make it obvious there is a problem so that we can investigate what the problem is and fix it
1 -
@Mormor192 The PID is helpful because then we can see exactly what you see and help with more specific answers. The PID is the set of numbers and letters near the name of the person whose profile you are viewing. You can click on it to copy it easily.
For example, the PID of one of my 3rd GGF:
0 -
In talking to counselors they always say ID, not PID, thus my unfamiliarity with that abbreviation/
0
