Change incorrect relationships from census
The 1940 census shows my great uncle as the father of my sister and was married to his niece. It should Zoe Eckelbarger was his niece and Laura Fay Ecklebarger was his nieces daughter not his
Also shows my father but no relationship to his wife and daughter.
I do show it correctly in my tree, but census record is wrong.
Answers
-
I believe this is the record @gloribell1 is referring to.
It's an interesting quandary regarding the transcription. It's technically been transcribed correctly, taking on face value the relationships have been recorded according to 'relationship to Head of Household' as the form intends.
However, it is evident on viewing the original that the census taker has taken som liberties with the format. First recording the head of household as Frank R Gordon, and noting his nephew-in-law (Harold I Eckelbarger) as that relationship corresponds to the 'Head'. But they have then recorded Zoe E Eckelbarger and Laura Fay Eckelbarger as they relate to Harold I Eckelbarger (wife and daughter), all under the same surname listing, also treating Harold as a 'Head' within the same household. Which has lead to the transcription issue.
Which begs the question, should it be read (and transcribed) literally in relation to the form, or as evidently intended by the census taker in the way they noted it down, ignoring the intended use of the relationship column. In this instance I think arguments could be made for both.
5 -
@RaniM - We're straying into philosophy here, rather than "how to" for FS, but misleading relationships are endemic in the UK censuses where 2 or more families exist in one household.
I believe that I can usually sort out who is a child of whom, but I am keenly aware that this is my interpretation and could be wrong.
That being so, if we can only have one transcript / index, it is my belief that it has to be the transcript of what is on the page. Any attempt at altering the index to "what it really means" is taking us further into danger. If the indexers / checkers are not from the same culture as the census, then sorry, but the risk is increased. You don't even need to go far - how many English people are liable to misinterpret a Scottish census because they fail to realise married women kept their maiden names? (Even I killed off an ancestor early because I forgot this!)
The only safe piece of advice is - look at the image, and cross check it with what you know (as the OP clearly did). If the image is there and someone doesn't do that check because they want to keep their stats up, they deserve all the trouble that they get.
5 -
North American indexers/users and FS AI also misinterpret a Scottish census, creating many unnecessary duplicates to sort out —— A big Thank You to @Adrian Bruce1 for explaining one reason why our family pedigrees can be so mixed up.
My English-Scottish friend tells me that Scottish Dads often gave their first son and daughter their mother's maiden name as a "middle" name so that if there was an inheritance to be had from mom's family, those children could legally claim the inheritance.
0 -
@Adrian Bruce1 I tend to agree with you. I have always been staunchly for transcriptions being to-the-letter exactly as they appear in the original. Hence my acknowledgement that technically the transcript is correct.
I have come across too many instances of people wanting to shoehorn changes to original records based on what they believe is accurate vs what the evidence suggests, to think it should be otherwise.
That being said, in this very specific instance I do find it an interesting case (ie unique and unusual), since it is very clear in the original that the census taker has recorded the relationships as they exist, but not as the pro forma dictates. Fortunately, as it is a US census record, the original can be easily citied to know this.
@gloribell1 This will have ramifications in the sourcelinker, which may be how you realised the issue to begin with. If you're wanting to attach the source to the relevant PIDs I suggest you manually override the preassigned relationships from the transcript using the "Change person" feature in the source linker, and just make a note in the "reason the source is attached" box, detailing the error as we've discussed here.
As far as UK census records are concerned, I don't think I've made it a secret my frustrations over some of the wild inaccuracies in the transcripts on FS (originating I believe from FMP), not just concerning relationships, but actual content regarding names, ages, places. Unfortunately, those originals aren't so readily available to cross reference or amend, which does lead to errors being propagated in the tree.
I agree that cultural practices play a role, and that's a very good point to highlight.
I would posit it would be more accurate to assert that Scots women retained their maiden name alongside, or interchangeably with, their married name. There are still plenty of instances where they do appear under their husband's name. (Bless the women who have been married two or three times and are recorded using every single surname plus their mother's maiden name and make my work a dream!)
3 -
@RaniM said
"... I would posit it would be more accurate to assert that Scots women retained their maiden name alongside, or interchangeably with, their married name. There are still plenty of instances where they do appear under their husband's name ..."
Yes, that would be a better way of putting it. I am totally unsure of what the current practice is, because I've never had to look. My impression is that the use of the husband's surname is typical now (I could be wrong) which implies a (presumably) gradual transition.
1 -
@maryellenstevensbarnes1 - I think that the inheritance reason for the use of surnames as Scottish middle names is probably over emphasized. Certainly there were inheritances depending on a name change but I suspect that in the majority of cases it's simply a desire to pay tribute to history. My GG-GF gave mostly different surnames as middle names to most of his children, including maiden names from prior to the mother's generation. At least one commemorated a family friend.
As with other things like Scottish naming patterns, some families followed very specific rules and others did the exact opposite. We have to investigate our own families with an open mind.
4 -
@Adrian Bruce1 and @RaniM
To clarify the issue of how married women are shown by surname in Scottish census returns, I have just checked through thirty pages I have previously downloaded and did not find one instance of a maiden name being used. (I looked at examples from 1841-1891 collections.) I believe the main area of a maiden name being recorded is that of civil death registrations, where the deceased woman's name appears to be usually recorded / indexed in both married and maiden names.
On the general issue being discussed here, I would agree that any census transcription should reflect what has been written in the original, regardless of how confusing that might be in line with what the enumerator might have intended to record, or what was the factual situation.
In establishing "true relationships", I find the situation particularly difficult where a second marriage (or cohabitation) has taken place. I do come across many records where the relationship to the head of the household is clearly shown: for example, "Step-son", in the case of a male child whose father was the wife's previous husband. However, I usually find a check of two of three collections (say for 1851,1861 &1871 E&W sets) is necessary (as well as reference to the GRO birth index) in order to establish the "facts" - as far as they can be determined, even when following that procedure!
Here is an example of what I encountered yesterday. Sorry if it appears to deviate from the OP's original issue (I'll come to that later!), but I think this represents the real problem in establishing relationships through census records. From the GRO index here is a list of the children of an Eliza Ann Haggett (maiden name):
(Name indexed / Year born / Mother's maiden name)
John Caesar Robinson 1890 (Blank)
William Riding Robinson 1892 Haggett
Thomas Riding Robinson 1895 (Blank)
Irvine Riding Robinson 1897 (Blank)
Mary Riding Robinson 1898 (Blank)
Albert Edgar Riding 1903 Haggett
The circumstances behind this are that Eliza HAGGETT married a Thomas ROBINSON in 1883, but they appear to have separated by 1890, when an apparently illegitimate John Caesar ROBINSON is born. He might (or might not have been) the biological son of Thomas RIDING, in whose household Eliza ROBINSON (status "Married") is found as housekeeper in the 1891 census, together with John C ROBINSON. However, by 1901 all members of the family are shown with the RIDING surname. And by 1911, Eliza and son William are shown as ROBINSON, but Albert as RIDING. ("Legally" this appears to accurately reflect the children's birth surnames, and Eliza's surname from her husband of 1883.) There is no record of Eliza ever marrying Thomas Riding and she disappears without trace by 1921 - when Thomas RIDING is found in the household of the wife and children of John Caesar ROBINSON - who had himself died in 1915, in WW1!
I believe this is an good example of how it can be extremely difficult (if not impossible) to work out accurate relationships through census records - even with the "back-up" of birth registrations. (I don't think I'm ever likely to establish whether John Caesar Robinson was the step or biological son of Thomas Riding.)
Finally, so as not to appear to completely disregard the subject originally raised, I wonder if @gloribell1 is asking for advice of how / whether to edit the relationships of the ECKLEBARGER family (as found in the 1940 census), or requesting "FamilySearch" to make edits that would make the relationships show as reflective of what they actually were? (The "rights and wrongs" of this have already been discussed in detail above, of course.)
5 -
@Paul W I have come across married women/widows using their maiden name in Scottish census records, but the majority I've researched have tended to be recorded under their married names, which aligns with your findings. I imagine different locations may also have some bearing on the practice being more or less prevalent. Civil death records are certainly one area they are (very helpfully) recorded. Maiden names are also used in the majority of children's baptism records, civil birth records, on some headstone inscriptions and interchangeably in wills.
Regarding OPs original question, the family in the census record are @gloribell1's immediate family, so they know definitively what the correct relationships are. I understood it to be more a question about editing the record to reflect that.
1
